Copyright
by
Tingji Tang
August 2006
The Dissertation Committee for Tingji Tang certifies that this is the approved
version of the following dissertation:
Photophysics of Perylene Diimides in Solutions and Self-Assembled
Films
Committee:
David A. Vanden Bout, Supervisor
Peter J. Rossky
Keith J. Stevenson
Alan Campion
Benny D. Freeman
Stephen E. Webber
Photophysics of Perylene Diimides in Solutions and Self-Assembled
Films
by
Tingji Tang, B.E.; M.S.
Dissertation
Presented to the Faculty of the Graduate School of
The University of Texas at Austin
in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
The University of Texas at Austin
August 2006
This work is for my wife Yanbo Wang, our son, Bryan Benshuo Tang, my daddy Jiaen
Tang, Mommy Chengfeng Lu and my brother Tingdong Tang.
Acknowledgements
I want to express my timeless gratitude to my research supervising Professor Dr.
Steve Webber, for his unselfish dedication to helping me through my studies and
research. I could not even begin to quantify the aid that he has given me. Both of his
moral and professional support has inspired me to be good scientist instead of a
technician. I am also very grateful to Dr. Vanden Bout for his generosity serving as my
official supervisor since Dr. Webber’s retirement. I really appreciate the collaboration
with Professor Klaus Müllen from Max-Planck-Institute for Polymer Research for his
wonderful compounds.
I would like to show my appreciation to various people who helped me in many
ways during my graduate school years. They are Dr. Mehdi Moini for his suggestions
concerning Mass Spectrscopy techniques; Dr. Yangming Sun for SEM traning and use;
Ms. Lynn Rozanski for her help in AFM measurements; Mr. Li-hsin Han for his
collaboration with me on the azo-related LBL study; Dr. Ti Cao for his intellectual
discussion of my research and fluorescence techniques; Dr. Jungseok Hahn and Mr. Ye
Hu for helping me in various ways in the labs;
I particularly want to thank my family: my parents, Jiaen Tang and Chengfeng
Lu and brother, Tingdong Tang. Thank you all for bringing me up with happiness and
love, which could not be more important. Lastly, by no means least, I wish to thank
Yanbo Wang for her endless love, sacrifice, understanding and encouragement. Bryan
Tang has become part of my life and happiness. I really enjoy the time we spent together,
not only past, but now, future and forever.
v
Photophysics of Perylene Diimides in Solutions and Self-Assembled
Films
Publication No._____________
Tingji Tang, Ph.D
The University of Texas at Austin, 2006
Supervisor: David A. Vanden Bout
New perylene diimide based dendrimers were synthesized by the divergent
synthetic method. Diamines with different lengths and methyl acrylate (or t-butyl
acrylate) were used as the extension regents for the dendrimer growth steps. The
photophysics of these synthesized PDIs was investigated with and without adding TFA.
Intramolecular fluorescence quenching was observed by comparing the fluorescence
intensity and lifetime before and after adding TFA, demonstrating a stronger quenching
process for those PDIs having 2 or 3 carbons between the two nitrogens of the diamines.
A 6 or 7 member ring mechanism for intramolecular electron transfer was presented.
Photophysics of new water soluble perylene diimide, anionic n-PDI, CAS
[694438-88-5] and cationic p-PDI, CAS [817207-4-7], was investigated in different
media. The absorption spectra and quantum yield were modified by the presence of these
reagents in a manner that suggests that weakly interacting complexes (such as H- or Jaggregates) are present in pure water and that complex formation can be enhanced by
vi
adding NaCl and polyelectrolytes or diminished by interacting with surfactants such as
SDS and DTAC.
A molecular layer-by-layer (MLBL) deposition process has been carried out for
n-PDI and p-PDI through the strong π−π and electrostatic interactions. The fluorescence
of these films shows an alternation of intensity according to which perylene species is on
the outer layer, which is interpreted as the effect of facile energy transfer between the
perylenes. Energy transfer in these films was also characterized by studying fluorescence
quenching produced by the deposition of a single energy trapping layer (n-TDI,
CAS[862852-56-0]). The MLBL technique can be utilized for other multiply-charged
water soluble conjugated dye molecules such as n-PSA (CAS[59572-10-0]) and n-TDI,
thereby demonstrating considerable potential for making organic semiconductor films.
Polyelectrolyte films containing n-PDI and p-PDI were prepared from PSS and
PDAC using the layer-by-layer (LBL) methodology.
The optical density and
fluorescence intensity of the LBL films grows linearly with the number of layers and we
do not observe extraction of the PDI by subsequent polyelectrolyte deposition in the
presence of 0.5M NaCl. The PDI fluorescence is substantially quenched in these films
which we interpret as a self-quenching effect, enhanced by inter- and intra-layer energy
transfer.
vii
Table of Contents
List of Tables……………………………………………………………………..…….xiii
Lsit of Figures……………………………………………………………………..........xiv
Chapter 1 Introduction………………………………………………………………….1
1.1
Peryelne Diimides…………………………………………………………1
1.1.1 Synthesis of Perylene Diimides…………………………………………..2
1.1.2 Applications of PDIs in Nanoscience and Nanotechnology……………...3
1.1.2.1 Organic Thin Film Transistors………………………………………5
1.1.2.2 Organic Light Emitting Diodes……………………………………...6
1.1.2.3 Photovoltaics………………………………………………………...7
1.1.2.4 Photoharvesting Arrays based on PDI dyes…………………………8
1.1.2.5 Supramolecular Self-Assembled Nanostructures……………………9
1.2 Layer-by-Layer Self-Assembly (LBL)………………………………………...10
1.3 References……………………………………………………………………..12
Chapter 2 Synthesis and Photophysics Study of 3, 4, 9, 10-Perylene
Tetracarboxylic Diimide Dendrimers…………………………………………22
2.1 Introduction………………………………………………………………..…..22
2.1.1 General Introduction…………………………………………………..….22
2.1.2 Nomenclature…………………………………………………………..…24
2.2 Experimental Section……………………………………………….................26
2.2.1 Materials……………………………………………………………..…...26
2.2.2 Instruments and Measurements………………………………………..….27
2.2.3 Synthesis of PDIs…………………………………………………………27
2.2.3.1 Synthesis of PDI dendrimers by Divergent method……………........27
2.3.1.1.1 Condensation of PDA with diamines…………………………...28
2.3.1.1.2 Michael Addition of PDI (0.0) with acrylates………………..…29
2.2.3.2 Synthesis of other PDIs by one step condensation…………………..30
2.3 Results and Discussions……………………………………………………….31
2.3.1 Absorption and Emission spectra of PDIs………………………………..32
viii
2.3.1.1 Molar Extinction Coefficient for PDIs…………………………..…..32
2.3.1.2 Emission Spectra of PDIs and Quantum Yield measurements…........33
2.3.2 Lifetime measurements for PDIs………………………………………….35
2.3.3 Photophysical behaviors of PDIs upon adding TFA…………...................35
2.3.3.1 Protocol for studies of the effect of added TFA…………………...…36
2.3.3.2 Photophysical properties of PDI compounds in the presence of TFA.36
2.3.3.2.1 Effect of TFA on the absorption spectra of PDIs……………...36
2.3.3.2.2 Effect of TFA on the fluorescence spectra of PDIs……………37
2.3.3.2.3 Lifetimes of PDIs in the presence of TFA……………………..39
2.4 Summary…………………………....................................................................39
2.5 References……………………………………………………………………..40
Chapter 3 Solution Photophysics of Water Soluble Perylene Diimides…………….68
3.1 Introduction……………………………………………………………………68
3.2 Experimental Section………………………………………………………….69
3.2.1 Materials………………………………………………………………….69
3.2.2 Instruments and Measurements…………………………………………...69
3.3 Results and Discussion……………………………………………...................70
3.3.1 Absorption and Emission Spectra of p-PDI and n-PDI in Solutions……..70
3.3.2 Photophysics of n-PDI and p-PDI in Surfactant Solutions.........................71
3.3.2.1 Photophysics Study upon complexing between PDI and appositively
charged surfactant (n-PDI/DTAC or p-PDI/SDS)……………………..74
3.3.2.2 Photophysics Study upon complexing between PDI and same charged
surfactant………………………………………………………………75
3.3.2.3 Photophysics Study upon Mixing the PDI/Surfactant complexes…….77
3.3.2.4 Photophysics study of n-PDI titrated by DTAC…………....................77
3.3.3 Photophysics study of n-TDI upon complexing with DTAC surfactant
solutions…………………………………………………………………….78
3.3.4 Effect of NaCl on the n-PDI and p-PDI Solutions…………………………78
3.3.5 Effect of PSS-Na+, PDAC+Cl- Polyelectrolytes on n-PDI and p-PDI
Solutions………………………………………..........................................79
ix
3.3.6 Molecular Dynamics Simulation of p-PDI Dimer…………………………81
3.4 Summary………………………………………………………………............82
3.5 References……………………………………………………………………..83
Chapter 4 Molecular Layer-by-layer Self-assembly of Water Soluble Dyes
through the π-π and Electrostatic Interactions…………………...............103
4.1 Introduction…………………………………………......................................103
4.2 Experimental Section……………………………………………...................104
4.2.1 Materials……………………………………………………………........104
4.2.2 Molecular Layer-by-Layer procedures…………………………………..105
4.2.3 Instruments and measurements…………………………………………..105
4.3 Results and Discussions………………………………...................................106
4.3.1 Molecular Layer-by-Layer of p-PDI and n-PDI………………………....106
4.3.1.1 Absorption spectra of (p-PDI/n-PDI)n……………………………....106
4.3.1.2 Characterization of (p-PDI/n-PDI)n…………………………….…...107
4.3.1.3 Emission spectra of (p-PDI/n-PDI)n………………………………...108
4.3.2 Molecular Layer-by-Layer of p-PDI and n-PSA………………………...109
4.3.2.1 Absorption spectra of (p-PDI/n-PSA)n……………………………...110
4.3.2.2 Emission spectra of (p-PDI/n-PSA)n………………………………..110
4.3.3 Molecular Layer-by-Layer of (p-PDI/n-TDI)n…………………………..111
4.3.3.1 Absorption spectra of (p-PDI/n-TDI)n………………………………111
4.3.3.2 Emission spectra of (p-PDI/n-TDI)n…………………………...........112
4.3.4 Molecular Layer-by-Layer of (p-PDI/(n-TDI:n-PSA))n…………………112
4.3.5 Energy Transfer for TDI/(p-PDI/n-PDI)n..................................................113
4.3.5.1 Absorption spectra for n-TDI/(p-PDI/n-PDI)n………………….......114
4.3.5.2 Energy Transfer for n-TDI/(p-PDI/n-PDI)n........................................114
4.4 Summary……………………………………..................................................116
4.5 References……………………………………………………………………118
Chapter 5 Incorporation of Water Soluble Perylene Diimides into
Poylelectrolyte Films by the Layer-by-Layer Method………………….133
5.1 Introduction…………………………………………………………………..133
x
5.2 Experimental Section………………………………………………………...134
5.2.1 Materials…………………………………………………………………134
5.2.2 Layer-by-Layer Procedures……………………………………………...134
5.2.3 Spin-Assisted Layer-by-Layer (SA-LBL)……………………………….135
5.2.4 UV-Vis and Fluorescence Spectroscopy Measurements……...................135
5.2.5 Atomic Force Microscopy (AFM) Measurements…………....................136
5.2.6 Förster Resonance Energy Transfer (FRET) Study……………………..136
5.3 Results and Discussions…………………………………………...................137
5.3.1 Dye Penetration into PEI/(PSS/PDAC)n LBL Films……………………137
5.3.2 LBL of n-PDI or p-PDI with PSS and PDAC…………………………...138
5.3.2.1 LBL with 0.5M NaCl in Polyelectrolyte Solution……………...…..138
5.3.2.2 LBL with no NaCl in Polyelectrolyte Solution……………………..141
5.3.2.3 Characterization of p-PDI or n-PDI in the LBL films………………143
5.3.3 Energy Transfer Study Between PDI films and BG……………………..144
5.4 Summary……………………..........................................................................146
5.5 References……………………………………………………………………148
Chapter 6 Some Preliminary Results and Suggestion for Future Work…………..166
6.1 Synthesis of n-type and p-type PDI materials………………………………..166
6.1.1 Introduction……………………………………………………………...167
6.1.2 Synthesis…………………………………………………………………167
6.1.2.1 Bromination of Perylene Dianhydride(PDA)……………………..167
6.1.2.2
Imidization of crude Bromination Product………………………..168
6.1.3 Photophysics of the synthesized PDIs………….......................................169
6.1.4. Future work……………………………………………………………..170
6.2 Self-assembly of n-type semiconductor PDIs………………………………..170
6.2.1 Introduction…….......................................................................................170
6.2.2 Self-assembly of PDIs…….......................................................................171
6.2.2.1 Chemical structures of PDIs………………………………………...171
6.2.2.2 Solution based Self-assemblying……………………………………171
6.2.3 Future of Self-assembly of PDIs………………………………………...173
xi
6.3 References……………………………………………………………………173
Bibliography……………………………………………………………...……………185
Vita……………………………………………………………………………………..187
xii
List of Tables
Table 2.1 Characterization of synthesized PDIs………………………………………....66
Table 2.2 Spectroscopic characteristic of PDI compounds with no TFA added…….......67
Table 2.3 Spectroscopic characteristics of PDI compounds with TFA added…………..68
Table 3.1 Photophysical parameters for n-PDI and p-PDI in different solutions………102
Table 5.1 FRET parameters for D-A pairs of p-PDI, n-PDI and BG…………………..164
Table 5.2 Characteristics of UV-Vis and fluorescence spectra of p-PDI and n-PDI LBL
films with 1, 3, and 5 SLs with and without using NaCl……………………………….165
xiii
List of Figures
Figure 1.1 Chemical Structures of Perylene and Peryelne derivatives………………....18
Figure 1.2 HOMO (top) and LUMO(bottom) of perylene diimides…………………....19
Figure 1.3 Procedure of synthesizing non symmetric PDIs……………………………..20
Figure 1.4 Schematic representation of Layer-by-Layer deposition process……………21
Figure 2.1 Schematic representation of Divergent and Covergent approach for dendrimer
synthesis………………………………………………………………………………….41
Figure 2.2 Scheme for the PDI compounds synthesis…………………………………...42
Figure 2.3 Mass spectrum of PDI(0.0)-16 by CI+………………………………………..43
Figure 2.4 Mass Spectra of PDI0.5MA-12 by CI+ and PDI0.5MA-13 by CI-…………...44
Figure 2.5 Mass Spectra of PDI0.5MA-2213 by CI (a) and PDI0.5tBA-12 by CI+ (b)....45
Figure 2.6 Mass Spectra of PDI0.5MA-14 (a) and PDI0.5MA-17(b) by CI-……………46
Figure 2.7 Mass Spectra of PDI0.5MA-112 by CI+……………………………………..47
Figure 2.8 Mass Spectra of PDI-DBPA (a) and PDI-18 (b) by CI+……………………..48
Figure 2.9 Chemical Structures of PDI compounds…………………………………….49
Figure 2.10 (a) Absorption spectra of PDI0.5MA-12 as a function of concentration. (b)
Absorbance at 526nm (●) and 490nm (▲) as a function of concentration for deriving the
molar extinction coefficient showing as the y value in the figure………………………51
Figure 2.11 (a) and (b) Normalized absorption (solid line) and emission (dashed line)
spectra for PDI0.5MA-112 and PDI-DBPA, respectively………………………………52
xiv
Figure 2.12 (a) Fluorescence quenching by the electron transfer mechanism from tertiary
amine. (b) Fluorescence electron transfer quenching mechanism suppressed by
combining with TFA…………………………………………………………………….53
Figure 2.13 6-member ring and 7-member ring quenching mechanism by electron transfer
from nitrogen to carbonyl group for PDI0.5-12, PDI0.5MA-13 and PDI0.5MA-2213,
respectively………………………………………………………………………………54
Figure 2.14 Fluorescence decay curve for PDI0.5MA-2214 before and after adding TFA.
Black: No TFA added, Red: adding TFA at [TFA]/[PDI]=106………………………….55
Figure 2.15 Absorption spectra responding to the addition of TFA for different PDI
compounds……………………………………………………………………………….56
Figure 2.16 Normalized emission spectra of PDI compounds at different [TFA]/[PDI]
ratios……………………………………………………………………………………...57
Figure 2.17 Ratio of fluorescence intensity to the absorbance at 490nm as a function of
Log ([TFA]/[PDI]). The red triangle represents the ratio where no TFA was added……62
Figure 2.18 The protonation of tertiary amine by TFA………………………………….63
Figure 3.1 Chemical Structures of n-PDI, p-PDI and n-TDI…………………………….87
Figure 3.2 Synthetic Scheme for n-PDI and p-PDI……………………………………...88
Figure 3.3 (a) and (b) Absorption and emission spectra of 1.5 x 10-6 M n-PDI and p-PDI
in water (solid line) and DMSO (dotted line), respectively……………………………..89
Figure 3.4 Absorption (left) and fluorescence emission (right) spectra in water for n-PDI
(black line), p-PDI (dark blue line), p-PDI/n-PDI mixed in a 1:1 ratio, (red line) and the
sum of the p-PDI and n-PDI spectra (light blue line)…………………………………..90
xv
Figure 3.5 Absorption spectra of n-PDI with DTAC (a) and p-PDI with SDS (b) at
different concentrations. …………………………………………..…………………….91
Figure 3.6 Fluorescence emission spectra of n-PDI (a) and p-PDI (b) mixing with SDS
and DTAC surfactant solutions at different concentrations, respectively……………….92
Figure 3.7 (a) and (b) Absorption and emission spectra of n-PDI upon mixing with SDS
solutions at different concentrations. Ex=540nm. ………………………………………93
Figure 3.8 (a) and (b) Absorption and emission spectra of p-PDI upon mixing with
DTAC solutions at different concentrations. Ex=540nm. ………………………………94
Figure 3.9 (a) Absorption and emission spectra of p-PDI/SDS mixing with n-PDI/SDS at
[SDS]=10mM (Red) and the average absorption spectrum of individual p-PDI/SDS and
n-PDI/SDS at [SDS]=10mM(black). (b) Fluorescence intensity upon mixing p-PDI/SDS
and n-PDI/SDS (red circle) and averaging the individual p-PDI/SDS and n-PDI/SDS
(black triangle) as a function of SDS concentration……………………………………..95
Figure 3.10
(a) and (b) Absorption and emission spectra of n-PDI titrated by
concentrated DTAC DMF solution. …..…………………………………………………96
Figure 3.11 (a) and (b) Absorption and emission spectra of n-TDI upon mixing with
DTAC solutions at different concentrations. Ex=640nm. ………………………………97
Figure 3.12 (a) Absorption spectra of n-PDI and (b) p-PDI as a function of NaCl
concentration, with insets showing the S-V fluorescence quenching curves……………98
Figure 3.13 Absorption spectra for n-PDI (a) and p-PDI (b) with the addition of PDAC
and PSS solutions.………………………………………………………………..………99
Figure 3.14 (a) Absorption and fluorescence spectra of n-PDI before (black) and after
(red) mixing with 1 mg/ml PDAC solution, demonstrating fluorescence quenching under
xvi
these conditions (b) Absorption and fluorescence spectra of p-PDI before (black) and
after (red) mixing with 1 mg/ml PSS solution, demonstrating fluorescence enhancement
under these conditions…………………………………………………………………..100
Figure 3.15 MD structure for p-PDI dimer in pure water………………………………101
Figure 4.1 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and n-PDI
(Red). (b) The peak absorbance around 575 nm as a function of monolayer PDI with pPDI as the first layer. (c) The peak absorbance around 284nm as a function of monolayer
PDI with p-PDI as the first layer……………………………………………………….120
Figure 4.2 The absorption spectra of p-PDI dipping cycles up to 10. ……..…………..121
Figure 4.3 (a) The TM-AFM image of (p-PDI/n-PDI)10 annealed at 80ºC overnight. (b)
Ellipsometric film thickness as a function of PDI monolayers………………………...122
Figure 4.4 (a) Fluorescence spectra from the monolayer of p-PDI/n-PDI with p-PDI as
the first layer (b) Fluorescence intensity as a function of double layer p-PDI/n-PDI….123
Figure 4.5 (a) Absorption and emission spectra of n-PSA in water with the inset showing
n-PSA’s chemical structure. Ex=360nm. (b) Absorption spectra of monolayer of p-PDI
and n-PSA starting with p-PDI layer. (c) Peak absorbance as a function of double layer
(p-PDI/n-PSA) for p-PDI at 575nm (circles) and n-PSA at 379nm (diamonds) after
correcting from the absorbance at 379nm for p-PDI…………………………………...124
Figure 4.6 (a) Emission spectra of 1st p-PDI (Red) monolayer and (p-PDI/n-PSA)1 (Pink).
Black curve is the subtraction of 1st p-PDI and (p-PDI/n-PSA)1. Ex=360nm. (b) Emission
spectra of (p-PDI/n-PSA)n-1/p-PDI with Ex=540nm. (c) Electron transfer mechanism
from the HOMO of PSA to p-PDI exciton……………………………………………..125
xvii
Figure 4.7 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and n-TDI
(Red). (b) The peak absorbance around 575 nm for p-PDI (black circles) and 700nm for
n-TDI (red diamonds) as a function of double layers. (c) The peak absorbance around
284nm as a function of deposition layers with p-PDI as the first layer………………...126
Figure 4.8 Emission spectra of (p-PDI/n-TDI)n with Ex=540nm………………………127
Figure 4.9 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and nTDI:n-PSA (Red). (b) The peak absorbance around 575 nm for p-PDI (black circles) and
700nm for n-TDI (red diamonds) as a function of the number of double layers……….128
Figure 4.10 Emission spectrum of (p-PDI/n-PDI)10(Red) and absorption spectrum of nTDI (Green)…………………………………………………………………………….129
Figure 4.11 (a) UV-Vis absorption spectra of n-TDI/(p-PDI/n-PDI)n. Green: n-TDI,
Black: p-PDI and Red: n-PDI. (b) Absorption spectra of n-TDI/(p-PDI/n-PDI)n upon
subtracting the absorption spectrum from n-TDI foundation layer. Black: p-PDI and Red:
n-PDI. (c) The peak absorbance around 579 nm as a function of monolayer PDI with pPDI as the first layer. …………………………………………………………………..130
Figure 4.12 (a) Schematic representation of deriving the average distance from (p-PDI/nPDI)4/p-PDI to the center of n-TDI layer. (b) Fluorescence emission intensity for (pPDI/n-PDI)n and n-TDI/(p-PDI/n-PDI)n as a function of PDI double layers. (c) Plot of
Log[(I0/Id)-1] versus Log(d), yielding a slope of 2.02 and intercept 3.61……………...131
Figure 4.13 (a) Energy transfer efficiency as a function of number of p-PDI layers. (b) pPDI fluorescence dependence as a function of number of p-PDI layers……………….132
Figure 5.1 (a) Absorbance at 580nm and (b) fluorescence intensity of n-PDI as a function
of dipping time for “precursor” LBL films of PEI/(PSS/PDAC)n (n=1, 2, and 3)……..133
xviii
Figure 5.2 (a) and (b) Absorption and fluorescence emission spectra of n-PDI dipping
into the preassembled PEI/(PSS/PDAC)6 as a function of immersing time……………151
Figure 5.3 (a) Absorbance at 587nm and (b) fluorescence intensity of p-PDI as a function
of dipping time for “precursor” LBL films of PEI/(PSS/PDAC)n-1/PSS……………….152
Figure 5.4 (a) and (b) Absorption spectra of n-PDI and p-PDI layer-by-layer assembly
with PSS and PDAC solutions made in 0.5 M NaCl and 3 SLs, respectively………….153
Figure 5.5 (a) and (b) Fluorescence emission spectra of n-PDI and p-PDI with PSS and
PDAC (0.5M NaCl) LBL films………………………………………………………...154
Figure 5.6 (a) Fluorescence intensity as a function of sequential deposition of
polyelectrolyte layers following the third p-PDI deposition (b) Fluorescence intensity as a
function of sequential deposition of polyelectrolyte layers following the third n-PDI
deposition………………………………………………………………………………155
Figure 5.7 (a) and (b) Absorption spectra of n-PDI and p-PDI with PSS and PDAC
(3SLs) in the LBL films (with no NaCl in the polyelectrolyte solution)………………156
Figure 5.8 (a) and (b) fluorescence spectra for n-PDI and p-PDI……………………...157
Figure 5.9 Absorption vs. the number of n-PDI layers deposited (●) and the normalized
fluorescence intensity (■) for each layer. The number of separation layers are indicated
between each n-PDI deposition (SL=1, 3, 5, 7, 9)…………………………………….158
Figure 5.10 TM-AFM angle view images of (a) before p-PDI deposition and (b) after pPDI depositions………………………………………………………………………...159
Figure 5.11 Schematic representation of p-PDI in LBL films with 3SLs along with
showing the intralayer and interlayer FRET process………………………………….160
xix
Figure 5.12 (a) and (b) Energy transfer efficiency as a function of spin-casting BG
concentration for 4 n-PDI and p-PDI LBL films with 3SLs……………………………161
Figure 5.13 (a) Energy transfer efficiency as a function of number of n-PDI layers at BG
concentration of 1.3x10-3M.
(b) Energy transfer efficiency as a function of number of p-
PDI layers at BG concentration of 8.3x10-6M………………………………………….162
Figure 5.14 Schematic representation of BG distribution in p-PDI and n-PDI LBL films
with 3SLs……………………………………………………………………………….163
Figure 6.1 Synthetic routes to Compounds PDI-NP1, PDI-NP2, and PDI-PN9……….176
Figure 6.2 Mass Spectrum of PDI-NP1 by CI+ (a) and H-NMR in CDCl3 (b)………...177
Figure 6.3 Mass Spectrum of PDI-NP2 by CI- (a) and H-NMR in CDCl3 (b)…………178
Figure 6.4. Mass Spectrum of PDI-PN9 by CI+………………………………………...179
Figure 6.5 Chemical Structures of Synthesized PDIs…………………………………..180
Figure 6.6 Normalized absorption (a) and emission spectra (b and c) of PDI-NP1 (Red),
PDI-NP-2(pink) and PDI-PN9 (Blue) in DCE………………………………………….181
Figure 6.7 SEM images of PDI-DEAPA……………………………………………….182
Figure 6.8 Large area SEM images of PDI-DMPA…………………………………….183
Figure 6.9 SEMs images of PDI-DMEA……………………………………………….184
xx
Chapter 1
Introduction
1.1 Perylene Diimides (PDIs)
Perylene is a conjugated planar molecule with 5 fused phenyl rings leading to
very strong intermolecular π-π interactions (see Figure 1.1a). Perylene dyes were first
discovered by Kardos in 1913 and have been used for a variety of applications as
pigments, including textiles, automobiles and general paints and plastics. Because of their
poor solubility due to strong π-π stacking interactions, the highly fluorescent nature of
perylenes diimides was not recognized until 1959. Since then, a variety of synthetic
schemes has been applied to modify perylene and perylene derivatives in order to
improve their solubilities and processibilities.
1
However, the synthesis of perylene
derivatives, such as perylene tetracarboxylic acid diimide (in short perylene diimide,
PDI), was not very successful until the discovery of 3,4,9,10-perylne tetracarboxylic
dianhydride (PDA) as shown in Figure 1.1b.
Since then, PDIs have received
considerable attention in academic as well as industrial dye and pigment research.
2
Perylene diimides were initially applied for industrial purposes as red vat dyes. After
1950, some PDIs were found to be high grade pigments (especially in automotive
finishes) due to their favorable combination of insolubility and migrational stability,
light- and weather-fastness, thermal stability and chemical inertness as well as high
tinctorial strength with hues ranging from red to violet, and even black shades.
3
Nowadays, PDIs have become very promising organic semiconductor materials for
various applications in novel high technologies such as photoconductors, filed effect
transistors, or photovoltaic devices.
1
1.1.1 Synthesis of Perylene Diimides (PDIs)
From the starting material of PDA, two synthetic routes have proved to be
successful for making perylene diimides (PDIs) derivatives (see Figure 1.1c) with high
solubility and functionality. Approach 1, discovered by Langhals et al in 1959, is to
introduce long chain alkyl groups at the imide nitrogen (head and tail positions) which
can dramatically increase the solubility of PDIs (see Figure 1.1c ). PDIs obtained by this
approach exhibit essentially indistinguishable absorption and emission properties because
the nodes in the HOMO and LUMO at the imide nitrogen reduce the coupling between
the perylene core and imide substitutents to a minimum, as shown in Figure 1. 2.
4
In
other words, PDIs can be regarded as a closed chromophoric system with an S0-S1
transition (polarized along the long molecular axis) whose intensity and position remain
unaltered by the specific imide substituents. Therefore, introducing long alkyl chains at
the head and tail positions does not affect the conjugation plane of the perylene core, so
that substituted PDIs usually have a high fluorescence yield and small Stokes Shift along
with well-resolved vibrational structures in both the absorption and emission spectra.
However, the ability to increase PDI’s solubility is limited due to the very strong π-π
interactions in solutions and solid films. Würthner et al at Universität Würzburg have
shown great success in synthesizing new functional materials via receptor substitutions at
the head and tail positions, which are then used as supermolecular assembly blocks. 5 The
second more elaborate synthetic strategy was to introduce substituents in the carbocyclic
scaffold in the so called bay-area (positions such 1, 6, 7, 12 in Figure 1.1c). This
approach was first discovered by Seybold and coworkers at BASF. They successfully
incorporated four phenoxy groups in high yields by nucleophilic displacement of chlorine
2
constituents, although the fourfold chlorination of PDA was contaminated by the threeand five- fold chlorination products.
6
Other nucleophilic displacements were not
successful for PDA chlorides. In 1997, di-substituent PDI derivatives were discovered
with the di-bromination of PDA.
7
However, the product mixture obtained by
bromination was even worse than in the case of chlorination because three-fold
bromination products and significant amounts of a second dibromo regioisomer are
formed which are only detectable in high field (>400MHz) H-NMR spectroscopy. Only
very recently was it discovered that the di-bromination of PDA was achieved in high
yields(>70%) with the addition of iodine at 80°C. 8 Even four-fold bromination of PDA
was obtained in a reasonable yield (30%) by adding additional iodine during reflux and
increasing the temperature and reaction time. 9 Gratifyingly, the displacement of bromine
substituents in PDA is very straightforward. Therefore, carbon, cyano, oxygen,
oligothiophene and nitrogen nucleophiles could be coupled to perylene core leading to
novel perylene diimides with unique properties.
10
In this case, the two naphthalene
planes that make up the perylene structure were twisted about 20-30 º depending on the
size and number of substituent groups. 5 Because of the easy control of the substituent at
the 1, 6, 7, and 12 positions, tetra- or di-substituted materials have been widely
synthesized. The introduction of polyphenylene dendrons at bay areas was systematically
carried out by Müllen et al at MPI.
11
The higher generation of polyphenylene dendrons
leads to diminished aggregation of PDIs in solutions and films due to their steric
hindrance. These dendrons can efficiently block the formation of PDI excimers and
exciplexes. Instead of small substituents at the bay positions, some star polymers with
perylene as the core have been synthesized and provide some unique mechanical
3
properties.
12
Therefore, bay-substitution is very versatile approach to control PDI
chemical structures. However, in all these cases, the absorption and emission spectra
were broadened and Stokes shift increases along with the decrease of quantum yield.
However, the solubility can be dramatically increased upon adding larger substituent
groups in the bay positions, due to the diminished π-π interactions.
While synthesizing symmetric PDIs which have the same groups at the nitrogen
positions is relatively straightforward by reaction with an excess of primary amine, nonsymmetric PDIs with different groups at the imide positions can not be prepared by a
stepwise condensation of primary amines with PDA. In some cases, a simultaneous
condensation of PDA with a mixture of primary amines has been carried out, but the
reactivity of the two amines must be similar and the reaction mixtures so obtained are
difficult to separate. Nagao, Misono, and coworkers have found a synthetic route by a
partial acid saponification of symmetric PDIs in conc. sulfuric acid at 180-200°C.
13
Simple aliphatic substituents R=methyl, ethyl, and butyl have been used. Problems occur
with aromatic substituents which can be sulfonated. An elegant synthesis has been found
by Tröster, which is shown in Figure 1. 3. 14 The insoluble PDA is converted to the tetra
potassium salt 2 which is readily soluble in water. The monoanhydride monopotassium
salt 3 is precipitated by a moderate acidification with orthophosphoric acid. The
procedure can be further optimized by the application of acetic acid, 15 because traces of
the latter can be more easily removed. The driving force for the formation of 3 is its
extraordinarily high lattice energy-the substance is absolutely insoluble in any solvent,
even at high temperature-which removes it from the protonation equilibria. 3 can be
condensed to nonsymmetric PDIs with primary amines in water. The necessity of an
4
aqueous medium, however, limits the scope of the condensation reaction. The yield is
very high for short chain water soluble amines, moderate with less soluble amines and
low with highly hydrophobic amines. However, the condensation of the latter is
important for the preparation of PDIs with a high solubility in organic solvents. 1
1.1.2 Applications of PDIs in Nanoscience and Nanotechnology
Molecules of PDIs form a unique class of n-type organic semiconductors
exhibiting relatively high electron affinity, brilliant colors, strong absorption and
fluorescence, and outstanding chemical, thermal and photochemical stability, in
comparison with the more common p-type counterpart in organic semiconductors. Bulk
phase materials have been widely used in various optoelectronic devices. Typical
examples include organic thin film transistors, photovoltaics, and organic light emitting
diodes (OLED). Due to their high thermal and photostability, PDIs have also attracted
increasing interest in fabrication of single molecule devices such as fluorescence
switches, sensors, molecular wires and transistors. Besides their applications as
semiconductors, they have been widely used in building blocks for making
supramolecular assemblies, nanobelts, and even photoharvesting systems through their
strong π-π intermolecular interactions.
1.1.2.1 Organic Thin Film Transistors (OTFTs)
Field-effect transistors (FET) devices, which can be prepared by both bottom-up
and top down methodologies, have been fabricated with thin films of perylene on SiO2
and polyimide gate insulators and p-channel FET properties have been found in both FET
devices, like the FET properties of pentacene and oligothiophenes. The hole mobility was
measured to be on the order of ~10-3 cm2V-1S-1 and current on/off ratios >105 were
5
obtained. 16 The highest hole carrier mobility achieved in p-type OTFTs was 1.5 cm2V-1S1
for pentacene transistors, which implies band transport of hole carriers. 17 Therefore, the
OTFTs based on n-type materials, in which electrons are the majority carrier, are also
desired since they will enable the fabrication of complimentary circuits. The “bottom-up”
approach requires the deposition of organic thin films on the pre-made substrates with
gate, source and drain electrodes on it while the " top-down” approach is to deposit the
organic thin films onto the gate substrate and then source and drain electrodes are coated
on top of the organic films. Malenfant et al. reported that the OTFTs of N,N’-dioctyl3,4,9,10-perylene diimide (PDI-C8) can reach an electron mobility up to 0.6 cm2V-1S-1
and current on/off ratios >105 using the bottom-up configuration.
18
Using Top-down
configuration, electron mobility of PDI-C8 OFETs made under higher vacuum (10-7 Torr)
can reach to 1.7 cm2V-1S-1 with on/off ratios around 107, agreeing with the trap-limited
band transport. 19
1.1.2.2 Organic Light Emitting Diodes (OLEDs)
Due to their high fluorescence quantum yield and excellent chemical and thermal
stability, PDIs are considered to very good materials for making OLEDs as the
emitting/electron transporting layer in OLED fabrications. For example, Jiang et al in
1997 reported the multilayered N, N’-bis(naphthyl)-3,4,9,10-perylene diimides based
OLED showing a maximum brightness of 26 cd/m2.
20
However, it exhibited self
quenching and strong interaction between the PDI layer and cathode metals. Since then,
the application of PDIs in OLEDs has suffered from two problems which reduce their
utility in such devices. First, many PDIs have low solubility, and therefore the
preparation of thin films layers has to be carried out by thermal vapor deposition, which
6
is very disadvantageous for fabricating large area films compared to spin-coating due to
its higher cost. Second, in the thin film solid state, PDIs have a very strong tendency to
form aggregates and even crystals, originating from the strong π-π intermolecular
interactions, which can lead to the decrease of emission efficiency due to the
enhancement of nonradiative decay pathways or red shifts in the emission. J. Qu, et al.
proposed an attractive way to solve both these problems simultaneously by incorporating
the PDI core within a polyphenylene dendrimer (one and second generation respectively)
shield (referred to as dendronized PDIs).
11
Polyphenylene dendrimers have the
advantage of being rigid and shape persistent and have been shown to have high chemical
and thermal stability. By connecting the polyphenylene dendrons of different generations
to a PDI core, the size of the resulting nanostructure can be controlled, and the PDI core
is separated from the surrounding media. The incorporation of polyphenylene dendrimers
at the bay positions is more efficient in shielding the π-π interactions especially at the
solid film states, also possessing excellent film formation properties. These dendronized
PDIs showed strong red-orange photoluminescence. However, single layer OLEDs of
dendronized PDIs showed low efficiencies and brightness and current density (0.14cd/m2
and 3.66mA/cm2 at 10V), especially with the second generation dendrimers, which is
attributed to the poor charge transport, presumably because the perylene diimide cores are
now too far apart. 11
1.1.2.3 Photovoltaics
Organic photovoltaic devices are potentially one of the most important emerging
new technologies, since the harvesting of energy by solar cells promises to be of
increasing necessity as our reserves of fossil fuels dwindle. The most widely used design
7
of organic solar cells are so-called bulk-heterojunction devices in which the active layer
consists of a blend of an electron-donating(p-type) and an electron-accepting (n-type)
material. Soluble polythiophenes and polyphenylenevinylene are widely used as the ptype materials. Employing PDIs as n-type materials in photovoltaics has gained much
attention due to their high electron affinity, large molar absorption coefficients and
possible generation of a highly conducting direction along the π-π stacking axis, in
addition to their other merits such as chemical robustness, thermal stability and low cost.
Tang C.W. at Kodak did the pioneering work of using PDIs as the active layer in solar
cells. 21 Solar cells fabricated with the bulk heterojunction from poly(2,7-carbazole) as the
p-type materials and PDI as the n-type materials showed a high external quantum
efficiency (EQE) of 16% at 490nm, FF around 0.4 and a power conversion efficiency of
0.6% illuminated with solar light. 22 The introduction of an electron donating group such
as pyrolidine at the bay positions of PDIs can lead to a 160nm bathochromic shift with
emission in the infrared region, which results in increasing the open circuit voltage (Voc)
and hence the current density due to the improved absorption overlap with the solar. 23
1.1.2.4 Photoharvesting Arrays based on PDI dyes
The development of efficient artificial systems for solar energy conversion is
important for sustainable energy utilization. Natural photosynthesis relies upon
monovalent interactions between similar chromophores to regulate energy and electron
flow. The π-π stacking in both solutions and the solid state can lead to the strong electron
wavefunction overlap between the adjacent PDIs, facilitating the energy and electron
flow upon photoexcitation. PDI based light harvesting arrays have been extensively
studied by Wasielewski in the last decade.
24
8
Hippius et al. have recently synthesized a
calix[4]arene-based light harvesting arrays containing
up to 5 PDI dyes (emitting in
orange, violet and green colors) in a cofacial arrangement.
Efficient energy transfer
could be observed even from initially nonfluorescent PDI to the green PDI moiety in the
molecule (fluorescence was quenched by photoinduced electron transfer). 25
1.1.2.5 Supramolecular Self-Assembled Nanostructures
Through precise molecular design one can obtain the spatial organization of PDIs
by molecular-recognition-directed self assembly. Such self-organization of functional
dyes should be of great utility to tailor defined multichromophoric objects or bulk solid
state materials with novel optical and electronic functionalities. π-π interactions, metalligand coordination, hydrogen-bonding and even ionic interaction can be combined to
direct the self-assembly of the individual PDI blocks in a defined manner. Würthner et al.
have done much of the research in the supramolecular chemistry in both solution and
solid state for the PDIs. 5
Although some organized structures (e.g., particles, liquid crystals, networks)
have been fabricated, 26 the 1D self-assembly (e.g. nanowires) of PDI molecules remains
challenging. Very recently, Zang Ling et al. used “phase transfer” self-assembly between
good and poor solvents and a surface annealing based self-assembly method for making
organized nanobelts of PDIs by controlling the π-π stacking and the hydrophobic
interactions between the side chains linked at the imide positions. In other words, the 1D
self-assembly of PDIs represents a balance between molecular stacking and solubility. 27
1.2 Layer-by-Layer Self-Assembly (LBL)
The preparation of organic thin films from the spontaneous assembly of functional
materials has been dramatically sped up by the introduction of the layer-by-layer (LBL)
9
technique introduced by Decher et al., based on the electrostatic interactions between the
polycations and polyanions.
28
The procedure is shown in Figure 1. 4 in which the
substrate is dipped alternatively into polycation and polyanion solutions. Two processes
for each polyelectrolyte dipping step are essential. The first is the charge compensation
and the other is the charge over-compensation which is crucial for the next
polyelectrolyte layer deposition.
29
LBL film fabrication can also be influenced by
secondary interactions such as hydrogen bonding, donor and acceptor interactions and ππ interactions.
30
Because of its simplicity, reproducibility and film uniformity over a
large area the LBL technique has been widely used in fabricating all kinds of heterostructural films incorporating dendrimers, quantum dots, polymer micelles, proteins and
even carbon nanotubes in support of applications such as chemical sensors, fuel cells,
nano- or ultra filtration, OLED etc.
29, 30, 31, 32
Besides producing flat films, the LBL
process can be conducted on substrates with other geometries such as polystyrene latexes,
optical fibers and even nanotubes. Free standing LBL films, hollow shells and hollow
nanotubes based on the LBL films can be obtained by the subsequent removal of the
substrates.
33
The mechanical properties of free standing LBL films were also
investigated with high modulus larger than 1GPa.c
34, 35
The recently discovered spray
assisted LBL process by Decher et al. can increase the time efficiency for making films
with desired compositions and thickness by 100 times. Spin-assisted LBL approach was
also discovered for making films more efficiently. 36
The film build up process can be monitored by all kinds of techniques such as
UV-Vis spectroscopy, ellipsometry, quartz crystal microbalance (QCM), Zeta potential,
neutron/X-ray scattering etc. Neutron reflectometry reveals the strong interpenetration of
10
the polyelectrolyte in the films up to 5 layers. 37 This also means that the polyelectrolyte
is not confined in a strict single layer fashion. Fluorescence Resonance Energy Transfer
(FRET) was also used to elucidate the nature of polyelectrolyte entanglement in the LBL
films.
38
A Free Energy Model for the LBL process has been developed by Park. et al
allowing for the predication of cumulative thickness as a function of number of deposited
layer as well as hydrated layer thickness and concentration of each absorbed layer within
the LBL films. 39
In addition to the use of polymeric molecules as the building blocks for
functional films, small dye molecules can be incorporated into LBL films, which
enhances their potential applications to OLEDs, photovoltaics and sensors. Furthermore,
incorporation of dye molecules into the films also enables one to use them as a probe to
obtain detailed information about molecular level local properties such as polarity or
molecular mobility. For example, pyrene molecules have been widely employed as the
molecular probes to sense the microenvironment polarity and viscosity.
40
A pyrene
derivative, 1,3,6,8-pyrene tetrasulfonic acid sodium salt, has been used in a LBL study of
both flat substrates and colloid particles. 41 Previous studies that applied the LBL process
to small charged organic dye molecules have shown that regular LBL film formation
depends critically on several factors: polymer concentration, ionic strength of the
polyelectrolyte solution and dye molecular shape and dimension.
42, 43
More recently,
LBL films combining small organic molecules with polyelectrolytes have been used as a
model system for drug release in which the assembled LBL films will decompose in
response to temperature or pH stimuli, thereby releasing the incorporated molecules.
44
The locally high density of small molecules in the LBL films has inspired researchers to
11
design the photo-harvesting architectures mimicking photosynthetic process.
45
However
it is often observed that the dye molecules are extracted by the next polyelectrolyte layer
(of opposite charge), which significantly impedes the loading efficiency of small
molecules into the films, affects the films’ integrity and stability and also compromises
the polyelectrolyte dipping solution. In the dissertation, we will demonstrate that this
extraction problem is minimal for our PDI molecules.
1.3 Reference:
1. Langhals, H. Heterocycles 1995, 40(1), 477-500.
2. a) Zollinger, H. Color Chemistry: Synthesis, Properties and Applications of
Organic Dyes and Pigments, 2nd, Rev. Ed. VCH: Weinheim, 1991. (b) Law, K. Y.
Chem. Rev. 1993, 93, 449.
3. Herbst, W. and K. Hunger, K. Industrial Organic Pigments: Production,
Properties, Applications, 2nd edn., WILEY-VCH, Weinheim, 1997.
4. H. Langals Spectrachim. Acta. 1988, 44A, 1189.
5. Würthner, F. Chem. Comm. 2004, 1564-1579.
6. (a) G. Seybold, G. and G. Wagenblast, G. Dyes Pigs. 1989, 11, 303-317. (b)
Sandrai, M.; Hadel, L.; Sauers, R. R.; Husain, S.; Krogh-Jespersen, K.;
Westbrook, J.D.; Bird, G.R. J. Phys. Chem. 1992, 96, 7988-7996.
7. Böhm, A.; Arms, H.; Henning, G.; Blaschka, P. (BASF AG), Ger. Pat. Appl., DE
19547209 Al, 1997.
8. Würthner, F., Stepanenko, V., Chen, Z., Saha-Möller, C. R., Kocher, N., Stalker,
D. J. Org. Chem. 2004, 69, 7933-7939.
9. Fan, L.; Xu, Y., He, T. Tetrahedron. Lett. 2005, 46, 4443-4447.
12
10. (a) Chao, C-C.; Leung, M-K., Su, Y.O., Chiu, K-Y.; Lin, T-H., Shieh, S-J., Lin,
S-C. J. Org. Chem. 2005, 70, 4323-4331. (b) Ahrens, A. H., Fuller, M. J.,
Wasielewski, M. R. Chem. Mater. 2003, 15, 2684-2686. (c) Zhao, Y.,
Wasielewski, M. R. Tetrahedron. Lett. 1999, 40, 7047-7050. (d) Lukas, A. S.,
Zhao, Y., Miller, S. E., Wasielewski, M. R. J. Phys. Chem. B. 2002, 106, 12991306. (e) Shen, S., Liu, Y., Qiu, W., Sun, X., Ma, Y., Zhu, D. Chem. Mater. 2005,
17, 2208-2215.
11. Qu, J., Zhang, J., Drimsdale, A. C., Müllen, K., Jaiser, F., Yang, X., Neher, D.
Macromolecules 2004, 37, 8197-8306.
12. Klok, H.A.; Becker, S.; Schuch, F.; Pakula, T.; Müllen, K. Macromol. Chem.
Phys. 2002, 203, 1106-1113.
13. Nagao, Y.; Misono, T. Bull. Chem. Soc. Jpn. 1981, 54, 1269.
14. Tröster H. Dyes Pigm. 1983, 4, 171.
15. Kaiser, H.; Linder, J.; Langhals, H. Chem. Ber. 1991, 124, 529.
16. Ohta, T.; Nagano, T.; Ochi, K.; Kubozono, Y.; Fujiwara, A. Appl. Phys. Lett.
2006, 88, 103506.
17. Lin, Y-Y.; Gundlach, D. J.; Nelson, S.F.; Jackson, T.N. IEEE Electron Device
Lett. 1997, 18, 606.
18. Malenfant, P. R.; Dimitrakopouls, C.D.; Gelorme, J. D.; Kosbar, L.; Graham, T.
O; Curioni, A.; Andreoni, W. Appl. Phys. Lett. 2002, 80, 2517.
19. Chesterfield, R.J.; Mckeen, J. C., Newman, C. R., Ewbank, P. C., da Silva Filho,
D. A., Brédas, J-L., Miller, L. L., Mann, K. R., Frisbie, C. D. J. Phys. Chem.B
2004, 108, 19281-19292.
13
20. Jiang, X. Z., Liu, Y. Q., Liu, X. G., Qiu, W. F., Song, X. Q., Zhu. D. B. Synth.
Met. 1997, 91, 253.
21. Tang, C.W. Appl. Phys. Lett. 1986, 48, 183-185.
22. Li. J.; Dierschke, F.; Wu, J. Grimsdale, A.C.; Müllen, K. J. Mater. Chem. 2006,
16, 96-100.
23. (a) Shin, W.S.; Jeong, H.-H.; Kim, M.-K.; Jin, S.-H.; Kim, M.-R.; Lee, J.-K.;
Lee, J. W.; Gal, Y.S. J. Mater. Chem. 2006, 16, 384-390. (b) Dinçalp, H., Içli, S.
Solar. Energy 2006, 80, 332-346.
24. (a) O’Neil, M. P., Niemaczyk M. P., Svec, W. A., Gosztola, D., Gaines III, G. L.,
Wasielewski, M. R. Science 1992, 257, 63-65. (b) Rybtchinski, B., Sinks, L. E.,
Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126, 12268-12269. (c) Ahrens, M.
J., Sinks, L. E., Rybtchinski, B. Liu, W., Jones, B. A., Giaimo, J. M., Gusev, A.
V., Goshe, A. J., Tiede, D. M., Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126,
8284-8294.
25. Hippius, C.; Schlosser, F.; Vysotsky, M.O.; Böhmer, V.; Würthner, F. J. Am.
Chem. Soc. 2006, 128, 3870-3871.
26. (a) Wang, W.; Li, L.-S.; Helms, G.; Zhou, H.-H.; Li, A.D.Q. J. Am. Chem. Soc.
2003, 125, 1120-1121. (b) Van, Herrikhuyzen, J.; Syamakumari, A.; Schenning,
A.P.; Weiss, E.A.; Wasielewski, M.R. J. Am. Chem. Soc. 2002, 124, 9582-9590.
(c) Sugiyasu, K.; Fujita, N.; Shinkai, S. Angew. Chem. Int. Ed. 2004, 43, 12291233.
27. (a) Balakrishnan, K., Datar, A., Naddo, T., Huang, J., Oikter, R., Yen, M., Zhao,
J., Zang, L. J. Am. Chem. Soc. 2006, ASAP. (b) Balakrishnan, K., Datar, A.,
14
Oikter, R., Chen, H., Zuo, J., Zang, L. J. Am. Chem. Soc. 2005, 127, 1049610497. (c) Datar, A., Oikter, R., Zang, L. Chem. Comm. 2006, 1649-1651.
28. (a) Decher, G.; Hong, J. Makromol. Chem. Symp. 1991, 46, 321. (b) Decher, G.
Science 1997, 277, 1232.
29. Tripathy, S., Kumar, J., Nalwa, H. S. Handbook of Polyelectrolytes and Their
Applications (V1): Polyelectrolyte-Based Multilayers, Self-Assemblies and
Nanostructures, 2002, American Scientific Publishers, California.
30. (a) Schmitt, J.; Decher, G.; Dressik, W. J.; Brandow, S. L.; Geer, R. E.;
Shashidbar, R.; Calvert, J. M. AdV. Mater. 1997, 9, 61. (b) Gao, M. Y.; Richter,
B.; Kirstein, S.; Möhwald, H. J. Phys. Chem. B 1998, 102, 4096. (c) Ariga, K.;
Lvov, Y.; Kunitake, T. J. Am. Chem. Soc. 1997, 119, 2224. (d) Lvov, Y.; Ariga,
K.; Ichinose, I.; Kunitake, T. J. Am. Chem. Soc. 1995, 117, 6117. (e) Caruso, F.;
Möhwald, H. J. Am. Chem. Soc. 1999, 121, 6039. (f) Tang, T.J.; Qu, J. Q.; Klaus,
M.; Webber, S. E. Langmuir 2006, 22; 26-28.
31. Decher, G., Schlenoff, J. B. Multilayer Thin Films, 2002, Wiley-VCH.
32. (a) Wang, Y.; Tang, Z. Y.; Correa-Duarte, M. A.; Liz-Marza’n, L. M.; Kotov, N.
A. J. Am. Chem. Soc. 2003, 125, 2830. (b) Wu, A.; Yoo, J.K.; Rubner, M.F. J.
Am. Chem. Soc. 1999, 121, 4883. (c) Hiller, L.; Mendelsohn, J. D.; Rubner, M. F.
Nat. Mater. 2002, 1, 59. (d) Tang, Z.; Wang, Y.; Kotov, N. A. Langmuir 2002,
18, 7035. (e) Park, J.; Hammond, P.T.; Adv. Mater. 2004, 6, 520. (f) Caruso, F.;
Caruso, R. A.; Möhwald, H. Chem. Mater. 2001, 13, 1076. (g) Harris, J.L.; Stair,
J.L.; Bruening, M.L. Chem. Mater. 2000, 12, 1941.
15
33. (a) Liang, Z., Susha, A. S., Yu, A., Caruso, F. Adv. Mater. 2003, 15, 1849-1853.
(b) Ai S., Lu, G., He, Q., Li, J. J. Am. Chem. Soc. 2003, 125, 11140-11141. (c)
Déjunat, C., Sukhorukov, G. B. Langmuir 2004, 20, 7265-7269. (d) Caruso, F.
Adv. Mater. 2001, 13, 11-22. (e) Li, Q., Quinn, J. F., Caruso, F. Adv. Mater. 2005,
17, 2058-2062.
34. (a) Markutsya, S., Jiang, C., Tsukruk, V. V. Adv. Funct. Mater. 2005, 15, 771780. (b) Jiang, C., Markutsya, S., Pikus, Y., Tsukruk, V. V. Nature Materials
2004, 3, 728. (c) Jiang, C., Markutsya, S., Tsukruk, V. V. Adv. Mater. 2004, 16,
157-161. (d) Mamedov, A. A., Kotov, N. A. Langmuir 2000, 16, 5530-5533. (e)
Ono, S. S., Decher, G. Nano. Lett. 2006, 6, 592-598.
35. (a) Nolte, A. J., Cohen, R. E., Rubner, M. F. Macromolecules, 2006, ASAP. (b)
Jaber, J. A., Schlenoof, J. B. J. Am. Chem. Soc. 2006, 128, 2940-2947.
36. (a) Izquierdo, A.; Ono, S. S., Voegel, J. C., Schaaf, P., Decher, G. Langmuir
2005, 21, 7558-7567. (b) Lee, S-S., Hong, J-D., Kim, C. H., Kim, K., Koo, J. P.,
Lee, K-B. Macromolecules 2001, 34, 5358-5360.
37. (a) Kellog, G. J., Mayes, A. M., Stockton, W.B., Ferreira, M., Rubner, M. F.
Langmuir 1996, 12, 5109. (b) Lösche, M., Schmitt, J., Decher, G. Bouwman, W.
G., Kajer, K. Macromolecules 1998, 31, 8893. (c) Tarabia, M., Hong, H.,
Davidov, D., Kirstein, S., Steitz, R., Neumann, R., Avny, Y. Appl. Phys. Lett.
1998, 83, 725.
38. (a) Kerimo, J., Adams, D. M., Barbara, P. F., Kaschak, D. M., Mallouk, T. E. J.
Phys. Chem. B 1998, 102, 9451. (b) Baur, J. W., Rubner, M. F., Reynolds, J. R.,
Kim, S. Langmuir 1999, 15, 6460. (c) Richter, B., Kirstein, S. J. Chem. Phys.
16
1999, 111, 5191. (d) Lowman, G. M., Daoud, N., Case, R. M., Carson, P. J.,
Buratto, S. K. Nano. Lett. 2001, 1, 677-682.
39. (a) Park, S. Y., Rubner, M. F., Mayes, A. M. Langmuir 2002, 18, 9600-9604. (b)
Schlenoff, J. B., Dubas, S. T. Macromolecules 2001, 34, 592-598.
40. Winnik, F.M. Chem. Rev. 1993, 93, 587.
41. (a) Tedeschi, C.; Caruso, F.; Möhwald, H.; Kirstein, S. J. Am. Chem. Soc. 2001,
122, 5841. (b) Caruso, F.; Lichtenfeld, H.; Donath, E.; Möhwald, H.
Macromolecules 1999, 32, 2317.
42. (a) Araki, K.; Wegner, M.J; Wrighton, M.S. Langmuir 1996, 12, 5393.; (b)
Cooper, T.; Campbell, A.; Crane, R. Lamgmuir 1995 11, 2713.
43. (a) Ariga, K.; Lvov, Y.; Kunitake, T. J. Am. Chem. Soc. 1997, 119, 2224.; (b)
Salomäki, M.; Tervasmäki, P.; Areva, S.; Kankare, J. Langmuir, 2004, 20, 3679.
(c) Linford, M.R.; Auch, M.; Möhwald, H. J. Am. Chem. Soc. 1998, 120, 178. (d)
Quin, J. F.; Caruso, F. Langmuir 2004, 20, 20.
44. (a) Wood, K.C.; Boedicker, J.Q.; Lynn, D.M.; Hammond, P.T. Langmuir 2005,
21, 1603-1609. (b) Yoo, P. J., Nam, K. T., Qi, J., Lee, S-K., Park, J., Belcher, A.
M. Hammond, P. T. Nature Materials 2006, 5, 234-240.
45. Dai, Z.; Dähne, L.; Donath, E.; Möhwald, H. J. Phys. Chem. B. 2002, 106, 11501.
17
R
O
O
O
3
3
5
1
3
5
1
6
12
7
11
8
10
9
(a)
4
2
5
1
6
Bay
Bay
7
12
6
12
O
4
2
4
2
O
Head
N
7
10
O
10
8
11
9
O
O
O
8
11
9
N
R
(c)
(b)
Figure 1.1 Chemical Structures of perylene and its derivatives
18
O
Tail
Figure 1. 2 HOMO (top) and LUMO(bottom) of perylene diimides.(Reproduced from
Würthner, F. 5)
19
O
O
O
O
CO2K CO2K
O
O
O
O
O
N
O
R-NH2
O
O
O
CO2K CO2K
CO2K CO2K
O
R
1
2
3
Figure 1. 3 Procedure of synthesizing non symmetric PDIs.
.
20
4
Figure 1.4 Schematic representation of Layer-by-Layer deposition
process.(Reproduced from Decher, G. et al.28b)
21
Chapter 2 Synthesis and Photophysics Study of 3, 4, 9, 10-Perylene
Tetracarboxylic Diimide Dendrimers
2.1 INTRODUCTION
2.1.1 General Introduction
Starburst dendrimers are monodisperse, hyperbranched polymers, which are
roughly spherical in shape and highly functionalized on the exterior. Because of their
well-defined, unique macromolecular structure, dendrimers are attractive scaffolds for a
variety of high end applications such as drug or gene delivery, light-harvesting, sensors,
and amplifications functions. 1-3
The synthesis of dendrimers has been achieved by the divergent and convergent
approaches, which are shown in Figure 2.1. Tomalia et al. disclosed the synthesis and
characterization of family of dendrimers in 1984-1985, using the so-called divergent
method. The synthesis was carried out by the Michael addition of a “core” molecule of
ammonia to three molecules of methyl acrylate, followed by exhaustive amidation of the
triester adduct that generates a molecule with six terminal amine groups. Iterative growth
is then continued using alternating Michael addition and amidation steps with the
appropriate excess of reagents.1 Since then, Newkome reported another family of
trisbranched polyamide dendrimers.2 Although dendrimers obtained by this approach can
be obtained on a commercial scale with molecular weight over 25,000, the repetition of
the coupling and activations steps leads to an exponential increase in the number of
reactions at the periphery (Figure 2.1a), which requires a large excess of reagents for each
step. Therefore, the likelihood of incomplete coupling or activation steps increases
exponentially as well. Although the removal of the monomer may be straightforward by a
22
simple distillation, precipitation or ultrafiltration, any flawed molecules resulting from
cyclization or incomplete reactions cannot easily be removed because of their structural
similarity to the intended product. In 1989-1990, Hawker and Fréchet introduced the
convergent growth approach to dendrimers with a more versatile and elegant control of
the dendrimer growth.3 This approach initiates growth from what will eventually become
the exterior of the molecule (Figure 2.1b), and progresses inward by coupling end groups
to each branch of the monomer (Figure 2.1b: coupling step). After completion of the
coupling, the single functional group located at the focal point of the wedge-shaped
dendron can be activated (Figure 2.1b: activation step). Coupling of this activated
dendron to each of the complementary functionalities on an additional monomer units
affords a higher generation dendron. After sufficient repetition of this process, these
dendrons can be attached to a multifunctional core through their focal point to form a
globular multidendron dendrimer. Although again an iterative synthesis, the convergent
route contrasts strongly with its divergent counterpart since it involves only a small
number of reactions per molecule during the coupling and activation steps. The
contribution of monomer to the mass of the product decreases exponentially as the
generation number increases. As a result, the chromatographic purification can be carried
out to produce the purest synthetic macromolecules prepared to date due to the large
dissimilarity between the monomers and dendrons.
This approach provides greater
structure control than the divergent approach due to its relatively low number of coupling
reactions at each growth step, allowing access to dendritic products of high purity and
functionalities. However, the commercialization is far more limited, because of it is less
readily scaled up than the divergent synthesis.
23
When compared with to linear analogues, dendrimers demonstrate significantly
increased solubility that can be readily tuned by the peripheral groups, and they also
exhibit very low intrinsic viscosities compared to linear polymers of comparable
molecular weight. As an alternative strategy to improve solubility and provide additional
chemical functionality for PDI’s, poly (amidoamine) (PAMAM) functionalized PDI had
been synthesized in our lab using PDA as the core through the divergent approach.4 From
these initial studies, it was found that there was obvious fluorescence quenching for all
these PDI dendrimers even in dilute solutions, probably the result of electron transfer
quenching from the tertiary amines in the PDI dendrimers. In this study, half-generation
PDI dendrimers with different length of alkyl chains in PDI-diimidediamine core were
prepared using the synthetic steps described in Figure 2.2. The photophysics of these
synthesized PDI dendrimers was studied in the presence trifluoroacetic acid (TFA),
which protonates the tertiary amine and prevents electron-transfer quenching. It is our
hope that the variety of our synthesized PDI dendrimers makes it possible to elucidate the
exact mechanism for the fluorescence quenching behavior.
2.1.2 Nomenclature
PDA=3, 4, 9, 10-perylene tetracarboxylic dianhydride.
PDI=3, 4, 9, 10-perylene tetracarboxylic diimide.
PDI (0.0)-1n=3, 4, 9, 10-perylene tetracarboxylic diimide core free amine (1,nalkyl diamine).
PDI0.5MA-12=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,2-ethylenediamine as Michael addition and diimide condensation
reagents, respectively.
24
PDI0.5tBA-12=3, 4, 9 ,10-perylene tetracarboxylic diimide dendrimer using tbutyl acrylate and 1,2-ethylenediamine as Michael addition and diimide condensation
reagents, respectively.
PDI0.5MA-13=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,3-propanediamine as Michael addition and diimide condensation
reagents, respectively.
PDI0.5MA-2213=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 2,2-dimethyl-1,3-propanediamine as Michael addition and diimide
condensation reagents, respectively.
PDI0.5MA-14=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,4-butanediamine as Michael addition and diimide condensation
reagents, respectively.
PDI0.5MA-16=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,6-hexanediamine as Michael addition and diimide condensation
reagents, respectively.
PDI0.5MA-17=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,7-heptanediamine as Michael addition and diimide condensation
reagents, respectively.
PDI0.5MA-112=3, 4, 9, 10-perylene tetracarboxylic diimide dendrimer using
methyl acrylate and 1,12-dodecanediamine as Michael addition and diimide condensation
reagents, respectively.
PDI18= N,N’-bis(1-aminooctadecane)-3, 4, 9, 10-perylene tetracarboxylic
diimide.
25
PDI-DBPA=N,N’-bis(N,N-dibutyl-3-aminopropyl)-3,
4,
9,
10-perylene
4,
9,
10-perylene
tetracarboxylic diimide.
PDI-DEPA=N,N’-bis(N,N-diethyl-3-aminopropyl)-3,
tetracarboxylic diimide.
PDI-DMPA=N,N’-bis(N,N-dimethyl-3-aminopropyl)-3,
4,
9,
10-perylene
9,
10-perylene
tetracarboxylic diimide.
PDI-DMEA=N,N’-bis(N,N-dimethyl-2-aminoethyl)-3,
4,
tetracarboxylic diimide.
PDI-DBP=Model Compound= N,N’-bis(2,5-ditert-butylephenyl)-3, 4, 9, 10perylene tetracarboxylic diimide.
2.2 Experimental Section
2.2.1 Materials
3,4,9,10-perylene tetracarboxylic dianhydride (PDA, 97% purity) was purchased
from Aldrich and used as received. All amines were purchased from Aldrich and used
without further purification. These amines were ethylene diamine, 1, 3-propanediamine,
2,2-dimethyl-1,3-propanediamine,
1,4-butanediamine,
1,6-hexanediamine,
1,7-
heptanediamine, 1,12-dodecanediamine and octadecylamine all with purity greater than
97%. Both methyl acrylate (MA) and t-butyl acrylate (t-BA) were obtained from Aldrich
and they were also used as received. PDI-DBP was obtained from Aldrich and purified
through flash chromatography by dichloromethane as the eluent. Trifluoroacetic acid
(TFA) was purchased from Fisher and used as received. Spectroscopy grade 1, 2dicholorethane (DCE) from Fisher was used for preparing PDI samples for the
26
photophysics study. The TLC plates we used were from J.T. Baker Baker-Flex Silica Gel
IB-F. Silicon gel was bought from Sorbent technology.
2.2.2 Instruments and Measurements.
Mass spectroscopy was taken on the ZAB2-E instrument by either negative or
positive chemical ionization.
UV-Vis absorption spectra were obtained with HP-8453 diode array spectrometer.
The fluorimeter used was a SPEX Fluorolog-τ2 equipped with a 450 W xenon
light source, Czerny-Turner double grating excitation and emission monochromators. A
photomultiplier voltage of 950V was typically used and the excitation and emission slit
widths were set at 0.5/0.5/1/1mm for solution samples. Emission spectra was collected in
front face mode with λex=490nm.
Time-correlated single-photon counting system (TCSPC) was used for the life
time measurements. A Mira 900-D Ti-Sapphire Mode-locked femtosecond oscillator
cavity with Optima Controller serves as the light source, whose tunable wavelength is in
the range of 700-980nm with pulse width less than 200fs. An optical frequency doubler
was coupled to extend the wavelength range from 350 to 490nm. The excitation
wavelength was adjusted at 463nm. The right angle mode was used for data collection.
The optical densities for all the samples were adjusted around 0.09. The data were
analyzed with the PicoQuant FluoFit Version4.0.
2.2.3 Synthesis of PDIs
2.2.3.1 Synthesis of PDI dendrimers by Divergent method
The general procedure of synthesizing the PDI dendrimers was described in
Figure 2.2. The first step is the imide condensation between PDA and diamines at high
27
temperature. The second step involves of Michael addition of either methyl acrylate or tbutyl acrylate.
2.3.1.1.1 Condensation of PDA with diamines
The synthesis was executed according to the procedures described in Dr. Cohen’s
dissertation 20 but with some improvements to making the synthesizing process easier and
more efficient. The general procedure was described in Figure 2.2. According to the
solubility of diamines, the condensation of different diamines with PDA can be divided
into in two categories. For example, 1,2-ethylenediamine, 1,3-propanediamine, 2,2dimethyl-1,3-propanediamine and 1,4-butanediamine are water soluble, therefore, the
purification of their condensation product with PDA can be achieved by precipitating the
reaction mixtures into water.
The amines with alkyl chains longer than 6 are not soluble
in water and also the amine reactivity is very low. Therefore, the reaction was carried out
in sealed tube (with three cycles of vacuum-melting-vacuum degassing) at the amine
refluxing temperature. After centrifuging the product solution, chloroform was used to
wash the solid until the pH paper did not turn blue indicating the complete remove of free
amines.
Protocol 1 for condensation of water soluble diamines with PDA. PDA (0.102 g,
0.26mmol) was mixed with 30 mL EDA (26g, 0.43mol) in a 250 mL flask with the aid of
35 mL pyridine. Before the flask was put into 100°C oil bath for refluxing under N2
overnight, the mixture suspension was ultrasonicated for 10 minutes. The final mixture
turned the characteristic deep purple color of the PDI compounds. The final solution was
centrifuged to remove the un-reacted PDA, and then poured into 350mL of water for PDI
(0.0) precipitation. The precipitation can be completely finished by leaving the mixture
28
overnight. Then the precipitate was collected by vacuum filtration using G4 filter funnel.
The product was dried overnight under high vacuum (~50milltorr). 0.1g PDI (0.0)-12 was
obtained with a yield of 95%. The product was verified by the CI(-)-MS.4b This product
was almost insoluble in any organic solvent. The 1H-NMR was performed on the formic
acid salt in D2O and showed the expected ratio of aromatic to aliphatic peaks.4
Protocol 2 for condensation of longer alkyl chain diamines with PDA. PDA
(200mg, 0.51mmol) was added into the tube and then 1,6-hexanediamine (0.6g,
5.17mmol) was added into it with the aid of heating. After three cycles of freezingmelting under vacuum, the tube was sealed under vacuum. The sealed tube was put into
120°C oil bath for 12 hrs. After 3 hours, the solution became dark purple color. High
vacuum was used to remove the excess 1, 6-hexanediamine with the liquid nitrogen trap.
Then chloroform was employed to wash the dark red solid until pH paper exposed to the
supernatant did not turn blue. The MS showed that this step was very successful without
byproduct, since there was only one strong peak at 589(m/Z, CI+), illustrated in Figure
2.3. This condensation was pretty successful (yield up to 94%).
By utilizing either protocol 1 or protocol 2, the PDI(0.0)s for different diamines
were synthesized and characterized by mass spectroscopy.
2.3.1.1.2 Michael Addition of PDI (0.0) with acrylates
According to Cohen’s initial study, the PDI (0.5) product has a very good
solubility in the co-solvent of CHCl3/CH3OH (2/1 by v/v).4 Therefore, the co-solvent of
CHCl3/CH3OH was used as the reaction media in order to make the reaction equilibrium
move to the product side. The Michael addition was carried out in a 100mL flask by
dissolving 50 mg PDI(0.0)-12 in 30 mL methyl acrylate (MA) with the aid of 15 mL
29
methanol and 30mL chloroform. Then the solution was refluxed at 60 °C under N2. TLC
was used to monitor the reaction process and the reaction was stopped after 3 days. The
final solution became deep red. After centrifuging, the methyl acrylate was removed by
rotavaporation with the aid of methanol. The yield for this process is less than 40%, due
to the incomplete Michael Additions for two-arm or 3-arm byproducts. The purified
product was obtained by silica-gel column chromography in CH2Cl2/CH3OH (95/05)
until only one point was observed on the TLC plate under UV-365nm exposure. Rf was
~0.42 in CH2Cl2/CH3OH (95/05). Only one strong peak (821 m/z, CI+) was found in the
CI-MS. Its mass spectrum was shown in Figure 2.4. Substituting t-butyl acrylate for MA,
PDI(0.5)t-BA-12 was then synthesized and purified by liquid chromatography following
the same steps as described above. Its mass spectrum by chemical ionization was shown
in Figure 2.5.
Following the above process, the other PDI(0.5) dendrimers were synthesized and
purified by liquid chromatography using the co-solvent CH2Cl2/CH3OH (95/05) as eluent.
Following the same protocol, PDI0.5MA-13, PDI0.5MA-2213, PDI0.5MA-14,
PDI0.5MA-17, and PDI0.5MA-113 were synthesized with yields between 10% and 40%,
whose mass spectra were shown in Figure 2.5 and Figure 2.6 and Figure 2.7. All these
compounds were characterized by TLC and CI-MS spectroscopy.
2.2.3.2 Synthesis of other PDIs by one step condensation
All other PDI compounds such as PDI18, PDI-DBPA, PDI-DEPA, PDI-DMPA
and PDI-DMEA were synthesized by the one step condensation between PDA and
primary amines.
30
For PDI-DBPA, PDA (0.3g, 0.76mmol) was added into 30mL pyridine. The
suspension was then refluxed in N2 for 20 minutes before 0.7mL 3-(dibutylamino)propylamine was added into it. The whole mixture was refluxed for 5 hrs, becoming deep
red eventually. After the reaction was stopped, the whole solution was precipitated into
acetone. After vacuum filtration, the dark red solid was obtained, which was weighed at
0.98g (yield about 92%). Its CI-MS is shown in Figure 2.7. The other PDIs were
synthesized following the same protocol and their MS is shown in Figure 2.8.
The molecular ion peak position, molecular weight and product yield of the
synthesized PDIs are listed in Table 2.1. As can be clearly seen, PDIs synthesized from
the one step condensation had higher yields. The Michael addition of acrylate to the
amine was not so effective because the strong π-π interactions from PDI (0.0) and
intermolecular cyclization etc.4
The chemical structures of all our PDI compounds are shown in Figure 2.9. The
differences in the chemical structure can allow us to study the structure-spectroscopy
relation of PDIs. Some PDIs have the different alkyl lengths in the spacer diamines, such
as ethylenediamine, propylenediamine, 1,4-butanediamine, 1,7-heptanediamine, and
1,12-dodecanediamine. PDI0.5MA-12 and PDI0.5tBA-12 have differences only in the
ester substituted group. The t-butyl groups are expected to have more steric hindrance for
more compact conformations. The branched structure in a diamine such as PDI0.5MA2213 was also investigated. The oxygen atom in the ester group from PDI0.5MA-13 will
also be studied compared with the similar PDI compound PDI-DBPA.
2.3 Results and Discussions
2.3.1 Absorption and Emission spectra of PDIs
31
2.3.1.1 Molar Extinction Coefficient for PDIs
Stock solutions were prepared in a 50mL volumetric flask at room temperature
using PDIs that were dried in vacuum at 70ºC overnight. For example, 155.91mg
PDI0.5MA-12 was dissolved in the 50mL dichloroethane (DCE) (molar concentration
1.9013x10-4 mol/L). This stock solution was left overnight before it was diluted into five
solutions with different concentrations Then the UV-Vis absorption spectrum of these
solutions was monitored, washing the UV cell by acetone between measurements.
The UV-Vis absorption spectra over a large range was monitored as shown in
Figure 2.10(a) and the absorbance at 526 nm and 490 nm were plotted against the
PDI0.5MA-12 concentration thereby deriving the molar extinction coefficients (ε). The
absorption spectra in DCE for PDI0.5MA-12 have a similar shape as the spectra for all
other PDIs in organic solvents. The broad S0-S1 transition along the long axis is located
between 450 nm and 540nm and the weak S0-S2 transition along the short axis shows a
second peak between 360nm and 450nm. A strongly pronounced vibronic structure was
observed with an absorption maximum at 526 nm which belongs to the electronic S0,0-S1,0
transition. The absorption has neither band shifting nor band broadening over our test
concentration range, which demonstrates that there is no strong intermolecular
interactions in this concentration range for PDI0.5MA-12. It is expected that the branched
structures in our PDIs impede the close contact of the inner perylene diimide cores of
adjacent molecules, preventing π-π stacking in the solution. The very strict linearity
between the absorbance for the two highest peaks and concentration suggests that this
expectation is correct. The molar extinction coefficients for the two highest bands were
found to be 76,000 and 48000 M-1cm-1 at 526 and 490nm, respectively. The fitting curves
32
were shown in Figure 2.10 (b). This is a little less than the reported value of 90,000 M1
cm-1for the model compound PDI-DBP.
Similar absorption spectra were obtained for all other synthesized compounds.
The highest two peak positions and their corresponding molar extinction coefficients are
listed in Table 2.2. As can be clearly see from Table 2.2, the peak positions are
independent of the chemical structure of all our synthesized PDI compounds and
essentially identical to the model PDI compound. This lack of spectral sensitivity is due
to the nodes present at the imide nitrogen positions (according to the MO calculations)
which make the electron coupling between the substituents at the head and tail positions
and the perylene core extremely weak.5 The molar extinction coefficient for all the
compounds lies in the range of 60,000 to 80,000 M-1cm-1 for the highest peak.
2.3.1.2 Emission Spectra of PDIs and Quantum Yield measurements
The normalized emission spectra reflect very good mirror images with respect to
the normalized absorption spectra for PDI0.5MA-112 and PDI-DBPA, as can be clearly
seen from Figure 2.11(a) and (b) respectively. The Stokes shifts for PDI0.5MA-112 and
PDI-DBPA are 12 and 11nm respectively. The Stokes shifts for other PDIs are listed in
Table 2.2. The relatively small Stokes shift indicates that the configurations and
interaction with the solvent of the excited and the ground state are very similar. The
quantum yields for all the synthesized PDIs compounds were measured in DCE using
PDI-DBP (QY=1) as the reference. The quantum yields for all the PDI compounds are
listed in Table 2.2. The quantum yields for PDI0.5MA-12, PDI0.5MAtBA-12, and
PDI0.5MA-2213 were 0.5 or less. All the other PDI compounds have QYs higher than
0.7.
PDI0.5-112 has a very high QY of 0.93 and PDI-18 has a QY of unity.
33
PDI0.5MA-2213 is anomalous in this series of compounds as it is the only compounds
with a three-methylene spacer that shows a relatively low fluorescence quantum yield,
and as we will see below, a strong enhancement of fluorescence in the presence of TFA.
Because it is an electron donor the tertiary amine nitrogen atom is good
fluorescence quencher by the electron transfer process shown in Figure 2.12. Due to the
electron donating properties of the alkyl chains in diamines, the tertiary nitrogen in the
PDI dendrimer is more prone to quenching the perylene diimide fluorescence. This is
why the QY of PDI18, without any tertiary amines, is unity. PDI0.5MA-112 has a much
higher QY than the other PDI0.5MA dendrimers, so we conclude that long distance
electron transfer along the alkyl chain is not possible nor is there a significant folding
back of the dendrimer “arms” onto the perylene core. Therefore we propose a quenching
mechanism based on the formation of a 6 or 7 member ring between the nitrogen atom
and carbonyl group in the PDI core, as represented in Figure 2.13. The higher electron
affinity for oxygen atom facilitates the electron transfer from electron rich nitrogen atom
to the electron deficient perylene core.
From the standpoint view of thermodynamic and
conformational analysis, 6 and 7 member rings are the most stable ones.6 In other words,
PDIs made from ethylenediamine and propylenediamide should have lower QYs due to
the ease of formation of 6 and 7 member rings. Even a bulky group like tert-butyl, as in
the case of PDI0.5tBA-12, cannot prevent the formation of rings. Evidently the branched
structure in diamines such as in PDI0.5MA-2213 cannot prevent the formation of
quenching rings either, as a very low quantum yield of 24% was found for PDI0.5MA2213. We presume this is due to the additional electron donating groups in the 2,2dimethyl-1,3-propylenediamine, which compensates for the larger N-O separation (i.e.
34
compare PDI0.5MA-13 and PDI0.5MA-2213). For other longer diamines the tertiary
amine in the PDI structure is separated from the perylene core by a larger distance and
cannot very easily form the required ring structure illustrated in Figure 2.13 although
thermodynamic fluctuations can bring the tertiary amine into closer proximity to the
perylene diimide chromophore.
2.3.2 Lifetime measurements for PDIs
The lifetimes for these PDI compounds in DCE are listed in Table 2.3. Only
monoexponential fluorescence decay was observed for PDI0.5MA-14, PDI0.5MA-17,
PDI0.5MA-112 and PDI-18 with a lifetime around 4ns, which is the typical lifetime for
PDIs.7 All the fluorescence decay curves were fitted with Σ2 less than 2.5 For the other
PDI compounds from the reaction with ethylenediamine or propylenediamine, a double
exponential fluorescence decay was observed with a fast decay component of
approximately 1 ns, which we assume corresponds to the electron transfer quenching
process from tertiary amine. For PDI0.5MA-2213, a more obvious fast component with a
decay lifetime of 0.23ns was obtained, which might be tentatively attributed to the high
electron density of amines and perhaps static quenching (discussed later). The
fluorescence decay curve of PDI0.5MA-2213 in DCE is shown in Figure 2.14. Two
longer lifetimes at 5.35 and 2.91ns were observed for PDI0.5tBA-12, which can be
attributed to the bulky structure of tertiary butyl groups in the exterior which makes
excited state relaxation more difficult.
2.3.3 Photophysical behaviors of PDIs upon adding TFA
It is well known that PDIs form aggregates easily in solution due to the π-π
interaction which leads to a diminished fluorescent quantum yield. The addition of an
35
acid such as TFA can potentially break up these aggregates and make the fluorescence
yield increase.8 Additionally, as a strong organic acid, TFA has very strong tendency to
protonate the tertiary amines which will greatly diminish electron-transfer quenching, as
illustrated in Figure 2.12(b). Thus the TFA effect on the fluorescence spectroscopy of
these PDI compounds was investigated and DCE was selected as the solvent because of
its low volatility.
2.3.3.1 Protocol for studies of the effect of added TFA
The TFA effect on the fluorescence spectra of PDIs in DCE was carried out using
the following process: Solutions of PDIs in DCE were prepared with a concentration of
approximately 10-6 M so that the optical density of the second highest was less than 0.1.
These PDI solutions were used to adjust the [TFA]/[PDI] ratio by progressive dilutions.
The molar concentration for pure TFA was calculated at 13.463 M. For each PDI
solution, a small amount of TFA was initially added into the above prepared PDI solution
so that [TFA]/[PDI] was 2x105. Then the prepared stock PDI solution was used to dilute
the above solution to make the second solution with [TFA]/[PDI] at 104. A third solution
was prepared by diluting the second solution to make [TFA]/[PDI] at 103 and so on until
the solution having [TFA]/[PDI] = 10-2 was prepared. Thus a series of PDI-TFA solutions
with the following ratios of [TFA]/[PDI] were obtained: 10-2,
102,
103,
104,
10-1,
100,
101,
2x105. All these solutions were sealed and placed in the dark
overnight before measuring their UV and fluorescence spectra.
2.3.3.2 Photophysical properties of PDI compounds in the presence of TFA
2.3.3.2.1 Effect of TFA on the absorption spectra of PDIs
36
Figure 2.15 demonstrates the change in the absorption spectra for some PDI
compounds upon adding TFA. As we can see, the absorption initially decreased to some
extent, then increased until the [TFA]/[PDI] exceeded 103. The decrease of molar
extinction coefficient for PDI0.5MA-12, PDI0.5MA-13, and PDI0.5MA-2213 is dramatic
compared with the other PDI compounds. For PDI0.5MA-112 and PDI18 there is hardly
any change in the molar extinction coefficients by the addition of TFA. In the literature, a
decrease in the molar extinction coefficient upon molecular aggregation is known as the
“hypochromic effect”.9 It is possible that the TFA association with the amines of the PDIs
might slightly perturb the PDI molecular orbital and therefore modify the extinction
coefficient also affected by the changing the solvent polarity and acidity. This can explain
the fact that no molar extinction change was observed for PDI-18 (which contains no
amine group) during the TFA titration. Clearly there is very strong red shift (~45nm) at
the highest TFA concentration examined. This is due to the fact that the energy of excited
state can be lowered by increasing the solvent polarity. However, this effect is not
obvious when the TFA amount is much smaller compared to the DCE solvent. This
results in the strong red shift only at very high TFA amount. The red shift for absorption
spectra at [TFA]/ [PDI] =2x105 is presented in Table 2.3.
2.3.3.2.2 Effect of TFA on the fluorescence spectra of PDIs
The normalized emission spectra at 4 different TFA concentrations for the PDI
compounds are shown in Figure 2.16. For each PDI compound a very structured emission
spectrum was observed with three major peaks corresponding to the transition from S1>S0, a mirror image of their absorption spectra. Similarly, a strong spectral red shift was
shown at a very high TFA concentration for each PDI. For a PDI compound such as
37
PDI0.5MA-12 (Figure 2.16a), PDI0.5MA-2213 (Figure 2.16b), PDI0.5tBA-12 (Figure
2.16c), PDI0.5MA-13 (not shown here), and PDI-DBPA (Figure 2.16f), there is a
progressive red shift with the increasing of TFA concentration. For PDI0.5MA-14
(Figure 2.16d), PDI0.5MA-17 (not shown here), PDI0.5MA-112 (not shown here) and
PDI-18 (Figure 2.16e), there is hardly any red shift before a very high TFA concentration
was reached.
We presume that the strong red shift at very high concentration is from the
high polarity of TFA. The volume fraction at [TFA]/[PDI]=2x105 is about 0.1.
In order to compare the exact effect of TFA on the fluorescence QY of PDI, the
normalized ratio of fluorescence intensity to the optical density at the excited wavelength
was plotted against the ratio [TFA]/[PDI], as shown in Figure 2.17 with the red point
indicating the value in the absence of TFA.
The ratio increases much more dramatically
for PDI0.5MA-12, PDI0.5MA-13, PDI0.5MA-2213, PDI0.5tBA-12 and PDI-DBPA,
whose QYs reached over 80% at [TFA]/[PDI]=4, compared with the other PDI cases. For
PDI0.5MA-14, PDI0.5MA-17 and PDI0.5MA-112, there is little change for the ratio after
a small jump with the initial addition of TFA (resulting in a QY higher than 90%). Once
again, at a very high TFA concentration, the QY started to drop. The smaller effect of
TFA on the QY for all these PDIs indicates that conformation fluctuations that can bring
the tertiary amine into closer proximity to the perylene diimide core are rare. For PDI18,
there is no change in the ratio until it begins to decrease at very high [TFA]/[PDI] ratio.
Upon the total protonation of the tertiary amines by TFA electron transfer from nitrogen
to perylene diimide through the carbonyls was suppressed (illustrated in Figure 2.18),
which leads to a larger jump around [TFA]/[PDI]=4 for the QYs of PDIs such as
PDI0.5MA-12, PDI0.5MA-13, PDI0.5MA-2213, PDI0.5tBA-12 and PDI-DBPA. For
38
other PDI compounds like PDI0.5MA-14, PDI0.5MA-17 and PDI0.5MA-112, there is
hardly any change in the ratio after the initial addition of TFA, suggesting that there is no
electron transfer quenching in DCE solution.
2.3.3.2.3 Lifetimes of PDIs in the presence of TFA
The lifetimes of PDIs at very high TFA concentration (TFA/PDI~106) were
measured and their values are listed in Table 2.2. A mono-exponentional decay of
fluorescence was detected for every PDI compound and all were fit by a lifetime around
4ns. The fluorescence decay curve for PDI0.5MA-2213 at a high TFA concentration is
shown in Figure 2.14 as the red curve. The simple exponential decay suggests that all
intramolecular quenching, including static quenching, has been suppressed. The lifetimes
for PDI0.5MA-14, PDI0.5MA-17 and PDI0.5MA-112 and PDI-18 showed not much
change before and after adding TFA. The lifetimes for PDI0.5MA-12, PDI0.5MA-13,
PDI-DBPA and PDI0.5MA-2213 became a little longer due to the suppressed quenching
from tertiary amine upon adding TFA. After adding TFA (especially at very high
concentration), the photophysics of the excited state of perylene core is about the same
for every compound and exhibit the same photophysics of relaxation to the ground state.
By comparing the two ratios of ΦTFA/ Φ0 and τ(TFA)/τ(DCE) (see Table 2.3), we found
that ΦTFA/ Φ0 is larger than τ(TFA)/τ(DCE) for PDI0.5MA-12, PDI0.5MA-2213 and
PDI0.5tBA-12, which means the presence of the static quenching for these compounds.
For all other compounds even for PDI0.5MA-13 and PDI-DBPA, these two values are
very close to each other indicating no static quenching. These static quenching can
further change the lifetime behavior of these compounds.
2.4 Summary
39
New perylene diimide based dendrimers were synthesized with an acceptable
yield, following the divergent synthetic method. Diamines with different lengths and
methyl acrylate (or t-butyl acrylate in one case) were used as the extension regents for the
dendrimer growth steps, leading to dendrimers with different spacers between the tertiary
amine and perylene diimide core. The photophysics of these synthesized PDIs was
investigated with and without adding TFA. The (presumed) intramolecular fluorescence
quenching was observed by comparing the fluorescence intensity and lifetime before and
after adding TFA, demonstrating a stronger quenching process for those PDIs that have 2
or 3 carbons between the two nitrogens of the diamines. The branched structure in the
diamine (PDI0.5MA-2213) leads to a more obvious quenching of fluorescence,
preseumably due to the higher electron density in the amine. Moreover, the static
quenching was observed for the compounds of PDI0.5MA-12, PDI0.5MA-2213 and
PDI0.5tBA-12 by comparing the ratios of ΦTFA/ Φ0 and τ(TFA)/τ(DCE). Upon
protonating the tertiary amine in the PDIs, all the PDIs show a similar single fluorescence
lifetime. By comparing the chemical structures of these synthesized PDIs (such as
PDI0.5MA-12 with PDI0.5tBA-12 and PDI0.5MA-13 with PDI-DBPA), we can
conclude that the ester groups in the PDI dendrimers did not have much effect on their
fluorescence behaviors. With the exception of PDI0.5-MA-2213, a spacer longer than
two –CH2- groups essentially eliminated intramolecular fluorescence quenching, a result
consistant with the ring-closure mechanism illustrated in Figures 2.13 and 2.18.
2.5 References:
1. Tomalia, D.A.; baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.; Roeck, J.;
Ryder, J.; Smith, P. Poly. J. 1985, 17, 117-132.
40
2. Newkome, G. R.; Yao, Z.; Baker, G. R.; Gupta, V.K. J. Org. Chem. 1985, 50,
2003-2004.
3. (a) Hawker, C.J.; Fréchet, J. M. J. J. Am. Chem. Soc. 1990, 112, 7638-7647. (b)
Grayson, S.M.; Fréchet, J. M. J. Chem. Rev. 2001, 101, 3819-3867.
4. (a) Cohen, T.S.; Parson, W.S.; Webber, S.E. Poly. Mater. Sci. Eng. 1997, 77, 228229. (b) Cohen, T. S. Syntheis and Photophysics of 3, 4, 9, 10-perylene
Tetracarboxylic Diimide Dendrimers, Ph.D. Dissertation, 2000. (b) Parsons, W.S.
Photophysics of Polymers in Solutions, The University of Texas at Austin, Ph.D.
dissertation, 1994.
5. H. Langals Spectrachim. Acta. 1988, 44A, 1189.
6. Organic Chemistry, 4th ed. John McMurry, Brooks/Cole Publishing, 1996.
7. Hofkens, J., Vosch, T., Maus, M., Köhn, F., Cotlet, M., Weil, T., Herrmann, A.,
Müllen, K., De Schryver, F. C. Chem. Phys. Lett. 2001, 333, 255.
8. Gvishi, R.; Reisfel, R.; Burshtein, Z. Chem. Phys. Lett. 1993, 213, 338.
9. El-Daly, S.A. Spectrochim. Acta A., 1999, 143.
41
Figure 2.1 Schematic representation of Divergent and Covergent approach for dendrimer
synthesis. (reproduced from Fréchet, J. M. J. 3b)
42
O
O
O
O
O
O
R-NH2
i
R-N
O
PDI
H2N-R-NH2
O
O
H2N-R-N
N-R-NH2
O
O
PDI0.0
ii
N-R
O
O
PDA
i
O
MA or t-BA
O
PDI0.5MA-12
PDI0.5tBA-12
PDI0.5MA-13
PDI0.5MA-2213
PDI0.5MA-14
PDI0.5MA-16
PDI0.5MA-17
PDI0.5MA-112
PDI-18
PDI-DBPA
PDI-DEPA
PDI-DMPA
PDI-DMEA
O
R'OOCH2CH2C
CH2CH2COOR'
N-R-N
N-R-N
R'OOCH2CH2C
O
CH2CH2COOR'
O
PDI0.5
Figure 2.2 Scheme for the PDI compounds synthesis.
43
R=1,2-ethylene, R'=methyl
R=1,2-ethylene, R'=t-butyl
R=1,3-propylene, R'=Methyl
R=2,2-dimethyl-1,3-propylene
R=1,4-butyl, R'=Methyl
R=1,6-hexane, R'=Methyl
R=1,7-heptane, R'=Methyl
R=1,12-dodecane, R'=Methyl
R=octadecyl
R=N,N-dibutyl-3-aminopropyl
R=N,N-diethyl-3-aminopropyl
R=N,N-dimethyl-3-aminopropyl
R=N,N-dimethyl-2-aminoethyl
Figure 2.3 Mass spectrum of PDI(0.0)-16 by CI+.
44
a
b
Figure 2.4 Mass Spectra of PDI0.5MA-12 by CI+ and PDI0.5MA-13 by CI-.
45
a
b
Figure 2.5 Mass Spectra of PDI0.5MA-2213 by CI- (a) and PDI0.5tBA-12 by CI+ (b).
46
a
b
Figure 2.6 Mass Spectra of PDI0.5MA-14 (a) and PDI0.5MA-17(b) by CI-.
47
Figure 2.7 Mass Spectra of PDI0.5MA-112 by CI+.
48
a
b
Figure 2.8 Mass Spectra of PDI-DBPA (a) and PDI-18 (b) by CI+.
49
O
H3COOCH2CH2C
H3COOCH2CH2C
O
CH2CH2COOCH3
NH2CH2C N
N CH2CH2N
O
CH2CH2COOCH3
O
PDI0.5MA-12
O
(H3C)3COOCH2CH2C
O
NH2CH2C N
(H3C)3COOCH2CH2C
CH2CH2COOC(CH3)3
N CH2CH2N
O
CH2CH2COOC(CH3)3
O
PDI0.5tBA-12
O
H3COOCH2CH2C
O
NH2CH2CH2C N
H3COOCH2CH2C
CH2CH2COOCH3
N CH2CH2CH2N
CH2CH2COOCH3
O
O
PDI0.5MA-13
H3COOCH2CH2C
H3COOCH2CH2C
H3C
O
O
NH2CCH2C N
H3C
CH3
N CH2CCH2N
O
O
CH3
CH2CH2COOCH3
CH2CH2COOCH3
PDI0.5MA-2213
H3COOCH2CH2C
O
O
N(H2CH2C)2 N
H3COOCH2CH2C
CH2CH2COOCH3
N (CH2CH2)2N
CH2CH2COOCH3
O
O
PDI0.5MA-14
50
O
H3COOCH2CH2C
H3COOCH2CH2C
NH2C(H2CH2C)3
O
N
N
O
(CH2CH2)3CH2N
CH2CH2COOCH3
CH2CH2COOCH3
O
PDI0.5MA-17
O
H3COOCH2CH2C
O
N(H2CH2C)12 N
H3COOCH2CH2C
CH2CH2COOCH3
N (CH2CH2)12N
CH2CH2COOCH3
O
O
PDI0.5MA-112
O
O
H3CH2CH2CH2C
CH2CH2CH2CH3
NH2CH2CH2CH2C N
H3CH2CH2CH2C
N CH2CH2CH2N
CH2CH2CH2CH3
O
O
PDI-DBPA
O
H3CH2C(H2CH2C)8
O
N
N
O
(CH2CH2)8CH2CH3
O
PDI-18
O
O
N
N
O
PDI-DBP
Figure 2.9 Chemical Structures of PDI compounds.
51
O
Absorbance
2.5
526nm
a
2
490nm
1.5
1
457nm
0.5
0
300
360
420
480
540
600
Wavelength(nm)
660
720
3.5
Absorbance
3
b
2.5
2
y = 75698x
1.5
R = 0.9997
2
y = 47834x
1
2
R = 0.9997
0.5
0
0.0E+00
1.0E-05
2.0E-05
3.0E-05
4.0E-05
PDI0.5MA-12 Concentration/M
Figure 2.10 (a) Absorption spectra of PDI0.5MA-12 as a function of concentration. (b)
Absorbance at 526nm (●) and 490nm (▲) as a function of concentration for
deriving the molar extinction coefficient showing as the y value in the
figure.
52
1.2
a
Norm. Abs.
1
1
Norm. Fluo. Inten.
1.2
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
350
400
450
500
550
600
650
0
700
Wavelength(nm)
1.2
b
Norm. Abs.
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
350
400
450
500
550
Wavelength(nm)
600
650
Norm. Fluo. Inten.
1.2
0
700
Figure 2.11 (a) and (b) Normalized absorption (solid line) and emission (dashed line)
spectra for PDI0.5MA-112 and PDI-DBPA, respectively.
53
P∗ +
NR 3
→
+
P− +
→
NR 3
P
+
NR 3
P−
e−
e−
→
→
(a)
P∗ +
NR 3
∗
+
TFA → P + H − NR 3
→
hν
→
P + H − + NR 3 + hν
→
(b)
Figure 2.12 (a) Fluorescence quenching by the electron transfer mechanism from tertiary
amine. (b) Fluorescence electron transfer quenching mechanism suppressed
by combining with TFA.
54
R
e
O
O
N
N
R N O
e
R
O
N
R
(a)
e
R
N
R
N
O
O
N
N
O
O
R
R
e
(b)
e
O
O
R
N
R
N
N
O
O
R
N
R
e
(c)
Figure 2.13 6-member ring and 7-member ring quenching mechanism by electron
transfer from nitrogen to carbonyl group for PDI0.5-12, PDI0.5MA-13 and
PDI0.5MA-2213, respectively.
55
Log[Intensity(counts)]
4
3
2
1
2
4
6
8
10
12
14 16 18
Time(ns)
20
22
24
26
28
Figure 2.14 Fluorescence decay curve for PDI0.5MA-2214 before and after adding TFA.
Black: No TFA added, Red: adding TFA at [TFA]/[PDI]=106.
56
0.12
0.12
0.11
a
0.1
0.09
Absorbance
0.1
0.08
0.08
0.07
0.06
0.06
-3
0.04
-2
-1
0.02
0
340
4
5
6
PDI0.5MA-12
400
460
0.12
520 580 640 700
Wavelength(nm)
760
820
880
0.11
b
0.1
Absorbance
0
1
2
3
Log([TFA]/[PDI])
0.09
0.08
0.07
0.06
0.05
-3
0.04
0.02
-2
-1
0
1
2
3
Log([TFA]/[PDI])
4
5
6
PDI0.5MA-13
0
340 400 460 520 580 640 700 760 820 880
Wavelength(nm)
0.1
c
Absorbance
0.08
0.06
PDI0.5MA-17
0.04
0.02
0
340
400
460
520
580
Wavelength(nm)
57
640
700
Absorbance
0.1
d
0.08
0.06
PDI0.5MA-112
0.04
0.02
0
340
Absorbance
0.06
400
460
520
580
Wavelength(nm)
640
700
e
0.04
PDI-18
0.02
0
340
400
460
520
580
Wavelength(nm)
640
700
Figure 2.15 Absorption spectra responding to the addition of TFA for different PDI
compounds.
58
Norm. Fluo. Inten.
1
a
0.8
0.6
PDI0.5MA-12
0.4
0.2
Norm. Fluo. Inten.
0
500
1
550
600
Wavlength(nm)
650
700
b
0.8
0.6
PDI0.5MA-2213
0.4
0.2
0
500
Norm. Fluo. Inten.
1
550
600
Wavelength(nm)
650
700
c
0.8
0.6
PDI0.5tBA-12
0.4
0.2
0
500
550
600
Wavelength(nm)
59
650
700
Norm. Fluo. Inten.
1
d
0.8
0.6
PDI0.5MA-14
0.4
0.2
0
500
Norm. Fluo. Inten.
1
550
600
Wavelength(nm)
650
700
e
0.8
0.6
PDI-18
0.4
0.2
0
Norm. Fluo. Inten.
500
1
550
600
Wavelength(nm)
650
700
f
0.8
PDI-DBAPA
0.6
0.4
0.2
0
500
550
600
Wavelength(nm)
650
700
Figure 2.16 Normalized emission spectra of PDI compounds at different [TFA]/[PDI]
ratios. Red: 0; Pink: 10; Black: 1000; Blue: 2x105.
60
Fluo. Inten./Abs(490nm)
1.4E+08
a
1.2E+08
1.0E+08
PDI0.5MA-12
8.0E+07
6.0E+07
-3
-2
-1
0
1
2
3
4
5
6
Fluo. Inten./Abs(490nm)
Log([TFA]/[PDI])
1.6E+08
b
1.3E+08
1.0E+08
PDI0.5tBA-12
7.0E+07
4.0E+07
Fluo. Inten./Abs(490nm)
-3
-2
-1
0
1
2
3
Log([TFA]/[PDI])
4
5
6
c
1.4E+08
1.3E+08
PDI0.5MA-13
1.2E+08
1.1E+08
-3
-2
-1
0
1
2
Log([TFA]/[PDI])
61
3
4
5
6
Fluo. Inten./Absorbance
1.6E+08
d
1.2E+08
8.0E+07
4.0E+07
PDI0.5MA-2213
0.0E+00
Fluo. Inten./Abs(490nm)
-3
-2
-1
0
1
2
3
Log([TFA]/[PDI])
4
5
6
1.6E+08
e
1.4E+08
1.2E+08
PDI-DBAPA
1.0E+08
-3
-2
-1
0
1
2
3
4
5
6
4
5
6
Fluo. Inten./Abs(490nm)
Log([TFA]/[PDI])
1.4E+08
f
1.2E+08
1.0E+08
PDI0.5MA-14
8.0E+07
-3
-2
-1
0
1
2
Log([TFA]/[PDI])
62
3
Fluo. Inten./Absorbance
2.0E+08
g
1.6E+08
1.2E+08
PDI0.5MA-17
8.0E+07
4.0E+07
Fluo. Inten./Abs(490nm)
-3
-2
0
1
2
3
Log([TFA]/[PDI])
4
5
6
4
5
6
4
5
6
h
1.7E+08
1.5E+08
1.3E+08
PDI0.5MA-112
1.1E+08
9.0E+07
-3
Fluo. Inten./Abs(490nm)
-1
1.9E+08
-2
-1
0
1
2
3
Log([TFA]/[PDI])
i
1.6E+08
1.3E+08
PDI-18
1.0E+08
-3
-2
-1
0
1
2
Log([TFA]/[PDI])
3
Figure 2.17 Ratio of fluorescence intensity to the absorbance at 490nm as a function of
Log ([TFA]/[PDI]). The red triangle represents the ratio where no TFA was
added.
63
R
O HN
O
N
N
R NH O
R
O
R
(a)
R
O
O
R
NH
N
N
O
O
NH
R
R
(b)
R
O
O
R
NH
N
N
O
O
HN
R
R
(c)
Figure 2.18 The protonation of tertiary amine by TFA
64
Table 2.1 Characterization of synthesized PDIs.
PDI0.5MA-12
PDI0.5tBA-12
PDI0.5MA-13
PDI0.5MA-2213
PDI0.5MA-14
PDI0.5MA-17
PDI0.5MA-112
PDI-DBPA
PDI-18
Yield
35%
28%
32%
16%
40%
12%
16%
92%
87%
MW(g/mol)
820
988
848
904
876
960
1101
728
895
65
MS Mode
CI+
CI+
CICICICICI+
CICI+
M+
821
990
848
904
878
960
1102
729
896
Table 2.2 Spectroscopic characteristic of PDI compounds with no TFA added
PDI0.5MA-12
PDI0.5tBA-12
PDI0.5MA-13
PDI0.5MA-2213
PDI0.5MA-14
PDI0.5MA-17
PDI0.5MA-112
PDI-DBPA
PDI-18
λ0,0-1,0
(nm)
526
528
527
524
524
524
524
524
524
ε0-0
(M-1cm-1)
76000
59000
60000
73000
62000
76000
74000
71000
90000a
λ0,0-1,1
(nm)
490
492
490
488
488
487
487
488
488
a
ε0-1
( M-1cm-1)
48000
42000
46000
49000
42000
48000
46000
45000
N/A
λ1,0-0,0
(nm)
533
534
533
535
532
532
531
533
531
Stokes
Shift(nm)
7
6
6
11
8
8
7
9
7
: Due to it low solubility, so it was assumed to have the same molar extinction as PDI-DBP.
66
Table 2.3 Spectroscopic characteristics of PDI compounds with TFA added
PDI0.5MA-12
PDI0.5tBA-12
PDI0.5MA-13
PDI0.5MA-2213
PDI0.5MA-14
PDI0.5MA-17
PDI0.5MA-112
PDI-DBPA
PDI-18
Red Shift a
9nm
7nm
9nm
12nm
12nm
12nm
11nm
13nm
11nm
Red shift b
18nm
15nm
16nm
14nm
18nm
18nm
18nm
18nm
17nm
Φ0 c
0.83
0.88
0.86
0.86
0.82
0.98
1.0
0.98
1.0
ΦTFAd
0.50
0.40
0.72
0.25
0.73
0.85
0.93
0.76
1.0
ΦTFA/ Φ0
1.66
2.2
1.19
3.44
1.12
1.14
1.07
1.29
1
a
: Ref shift of λ0,0-1,0 with adding TFA at [TFA]/[PDI]=2x105.
: Ref shift of λ1,0-0,0 with adding TFA at [TFA]/[PDI]=2x105.
c
: Quantum yiled of PDI with adding TFA at [TFA]/[PDI]=2x105.
d
: Quantum yiled of PDI without TFA.
e
: lifetime of PDI without adding TFA in DCE.
f
: lifetime of PDI with adding TFA at [TFA]/[PDI]=2x105.
g
: The average lifetime was taken for calculation.
h
: The number in the parenthesis is the pre-exponential factor.
b
67
τ(DCE)e
3.89(0.70h),1.24(0.30)
5.35(0.31), 291(0.69)
3.86(0.85), 1.24(0.15)
3.83(0.46), 0.23(0.52)
3.81
4.01
4.04
3.71(0.84), 0.92(0.16)
4.07
τ(TFA)f
4.07
4.16
4.20
4.15
4.18
4.12
4.22
4.17
3.97
τ(TFA)/τ(DCE)g
1.31
1.13
1.21
2.21
1.10
1.04
1.04
1.28
0.98
Chapter 3 Solution Photophysics of Water Soluble Perylene Diimides
3.1 Introduction
In
recent
years,
different
attempts
were
made
to
improve
the
photophysical/photochemical properties of the water-soluble fluorescent dyes as labels
for cells, antibodies, and DNA, which has gained additional momentum owing to the use
of single molecular spectroscopy to gain deep insight into biological systems.1 Although
there are a large variety of water-soluble chromophores commercially available today,
most of them exhibit relatively low fluorescence quantum yield and/or photochemical
instabilities. Cyanine dyes, for example, are highly unstable towards oxygen and light,
while xanthene dyes tend to aggregate in aqueous medium. 1a, 2 The ideal dye should have
the following characteristics: 1) good water solubility, 2) high fluorescence quantum
yield, 3) high chemical and photostability, 4) nontoxity, 5) good biocompatibility, as well
as 6) possible commercial viability and scalable production.
The highly fluorescent 3, 4, 9, 10-perylene tetracarboxylic diimide (PDI)
chromophore is widely used as a commercial dye and pigment due to its outstanding
chemical, thermal and photochemical stability. Owing to their optical properties PDIs
should be excellent biological probes.3,
4
However, water soluble PDIs with
functionalities at their imide sites showed poor water solubility and very weak
fluorescence in water due to the strong π-π stacking.5 It is well known that the solubility
of PDIs in organic solvents can be dramatically increased when the substituents are
attached in the bay region.6 It is expected that water solubility can be achieved by
introducing charged groups in the bay region which also results in a twist within the
perylene core leading to less tendency for π-π stacking. Following this strategy, Qu J. et
al. synthesized water soluble PDIs with moderate to high fluorescence quantum yields
68
which have demonstrated their promise as the biological labels for living cells.4,6
However, their solution photophysics of these water soluble PDIS have not been
systematically studied. In this chapter, the photophysics of water soluble PDIs in different
media was investigated.
3.2 Experimental Section
3.2.1 Materials
Positively and negatively charged water soluble p-PDI (CAS [817207-4-7]) and nPDI (CAS [694438-88-5]) (Figure 3.1, with dimensions estimated from MD simulations
of 2.13x1.33x1.89 nm and 2.13x1.15x1.49 nm (LWH), respectively 7) were synthesized
by Qu, J. et al with synthetic scheme presented in Figure 3.2.
Poly(diallydimethylammonium chloride) (PDAC, Mw = 25,000g/mol), and
poly(styrene sulfonate) (PSS, Mw = 70,000g/mol) were purchased from Sigma-Aldrich
and used without further purification. Sodium dodecyl sulfate (SDS) was purchased from
Aldrich and crystallized in ethanol before use. Dodecyltrimethylammonium chloride
(DTAC, puriss grade ≥99%) was purchased from Fluka and used as received. The anionic
and cationic surfactants SDS and DTAC were employed in these studies (their
corresponding CMC are 8 and 20mM in neutral water, respectively). The effects of
surfactants such as SDS and DTAC or NaCl on the absorption and fluorescence emission
spectra of p-PDI and n-PDI were studied by diluting concentrated p-PDI and n-PDI
aqueous solutions with the SDS and DTAC solutions of different concentrations. All
samples were stirred overnight before taking their spectra.
3.2.2 Instruments and Measurements
Same UV-Vis spectrometer and Spex τ2 were used for measurements except that
the Ex/Em slit width used in these experiments were 1/1/2/2 mm.
69
Same TCSPC instruments were used to get the lifetime of p-PDI and n-PDI in
different medium except that excitation light source at 463nm was used instead at 422nm.
3.3 Results and Discussion
3.3.1 Absorption and Emission Spectra of p-PDI and n-PDI in Solutions.
Due to strong intermolecular forces, such as π-π interaction, self-association of
dyes in aqueous solution is often encountered and similar behavior would be expected for
n- and p-PDI.
It is to be expected that in DMSO water soluble PDIs will remain in
monomeric form, as has been found for larger π-electron system such as water soluble
terylenediimdes.8 In Figure 3.3 is shown the absorption and emission spectra of p-PDI
and n-PDI in water and DMSO, respectively. Their absorption spectra in water have a
similar shape as the spectra for other PDIs in organic solvents.9 The broad S0-S1 transition
along the long axis is located between 480 nm and 630nm and the weak S0-S2 transition
along the short axis shows a second peak between 360nm and 480nm. For n-PDI in water
the S0-S1 transition has a maxima peak at 568nm (λ0-0; ε=2.97 x 104 M-1cm-1) and a
shoulder at 540nm, and its emission spectrum is composed of only one peak at 618nm
with a quantum yield of about 0.54 and a large Stokes shift of 50nm, as in Figure 3.3a.
When n-PDI was dissolved in DMSO the absorption spectra was very similar to that
reported for other PDIs in the monomeric form,
21
showing a better-resolved vibrational
structure with maxima at 577nm (λ0-0; ε=3.44 x 104 M-1cm-1) and 540nm (λ0-1, ε=2.26 x
104 M-1cm-1). The increase of the molar extinction coefficient in DMSO is attributed to
breaking up of n-PDI aggregates.10 The emission peak in DMSO was centered at 611nm
with the quantum yield increasing from 0.54 in water to 0.82 in DMSO. The Stokes shift
was decreased to 34nm in DMSO.
70
A similar trend was observed for p-PDI in water and DMSO, except that both
absorption and fluorescence spectra are red shifted in water (λ0-0 was red shifted from
581nm (3.85 x 104 M-1cm-1) in DMSO to 587nm (3.28 x 104 M-1cm-1) in water)
suggesting the formation of aggregates in water, as shown in Figure 3.3b.
5,7,11
The
quantum yield in DMSO was three times larger than in water. Again, the increases in
both quantum yields and molar extinction coefficient for p-PDI in DMSO suggest the
presence of monomeric form of p-PDI in DMSO. The Stokes shift was about the same as
in these two solvents.
Spectral data are collected in Table 3.1.
According to Molecular Exciton Theory, two main types of dye aggregates are
distinguished as H- and J- aggregates.11 H-aggregates are formed with the face-to-face or
card-pack stacking configuration, leading to the absorption blue shift in respect to the
isolated chromophore and a total loss of fluorescence if two PDI’s transistion dipoles are
strictly parallel to each other. However, J-aggregates with the head-to-tail configuration
usually shows fluorescence featured by the red shifts of both absorption and fluorescence
emission spectra peaks.11 Comparing these cases with our observations (see Table 3.1),
we conclude that p-PDI forms H-aggregates in water. However, the orientation of dipole
moments from the adjacent p-PDI molecules are most likely not strictly parallel to each
other, according to MD simulations of the p-PDI dimer. For n-PDI, although it had a blue
shift of absorption peak in water by 9nm relative to that in DMSO, it has a very strong
emission from water with a red shift of about 7nm. Therefore, we assume that the n-PDI
forms aggregates whose configuration lies between H- aggregates and J-aggregates.
Lifetimes of n-PDI and p-PDI in both water and DMSO are listed in Table 3.1.
All the lifetimes were fit with Σ less than 2. As can be seen from Table 3.1, fluorescence
decays for n-PDI was monoexponential with a lifetime of 4.78 and two lifetimes were
71
obtained for p-PDI in water (6.35 and 0.13 ns) showing that there probably exist rather
complex structures of p-PDI in solution. The very short lifetime species might be
attributed to the H-type like aggregate of p-PDI in water (see above or later simulation).
Slightly longer lifetimes of n-PDI and p-PDI at DMSO, following mono-exponential
decay, were observed at 5.49ns and 5.41 ns, indicating the dissociation of n-PDI and pPDI aggregates in water. The fact that the fluorescence lifetimes in water and DMSO are
fairly similar but there is an appreciable difference in fluorescence quantum yield is
consistent with the presume of dimers or aggregates and static quenching of fluorescence
for p-PDI. The following equations are usually applied to extract the the fluorescence rate
constant (kf ) and nonradiative rate constant (knr):
φf =
τ=
kf
(1)
k f + k nr
1
k f + k nr
(2)
For example, n-PDI in DMSO has a lifetime of 5.49ns and quantum yield of 82%,
we obtain kf=1.49x108s-1 and knr=3.31x107s-1. Hence, kf>knr. The other kf and knr values
are listed in Table 3.1. A higher fluorescence rate constant was obtained for n-PDI and a
higher nonradiative rate constant for p-PDI.
In order to examine the complexation between the p-PDI and n-PDI, experiments
in solution were conducted. Both the absorption and fluorescence spectra were shown in
Figure 3.4 for p-PDI, n-PDI, and the p-PDI/n-PDI mixture (1/1 mol/mol), respectively in
water at pH=6.5. The simple sum of the p-PDI and n-PDI UV individual absorption and
fluorescence spectra are also displayed in Figure 3.4 labeled as “p-PDI + n-PDI”. UV
spectra for p-PDI/n-PDI (1/1) and p-PDI + n-PDI are not identical to each other, which
suggests that a complex is formed between the p-PDI and n-PDI upon mixing.
72
However, the absorption spectrum of the 1:1 complex is only slightly modified from the
individual species, unlike most examples of PDI aggregates.12 The oppositely charged
PDI molecules favor the formation of p-PDI and n-PDI complexation through both the
electrostatic and π-π interaction. The fluorescence from the 1/1 PDI mixture is much
lower than p-PDI, n-PDI or p-PDI + n-PDI due to the formation of a complex
(presumably via π-π stacking) between the p-PDI and n-PDI. The complexation of p-PDI
and n-PDI with a 1:1 ratio have the following optical characteristics: absorption peaks at
450 nm, 545 nm and 577 nm and an emission peak at 614 nm. At the same time, the
emission from p-PDI/n-PDI is about 7.6% of that of p-PDI, or a quenching efficiency of
approximately 13.
3.3.2 Photophysics of n-PDI and p-PDI in Surfactant Solutions.
The Ionic Self-Assembly (ISA) process utilizes the electrostatic interactions
between charged surfactants and oppositely charged oligoelectrolytic species as the
primary interaction.13 Other noncovalent interactions such as H-bonding, π-π and
hydrophobic interactions can also promote this self-assembly process.14 By selecting
different functional moieties, ISA materials can be tailored to show desired properties
such as switchable luminescent
and reversibly switchable conducting properties.15
Only a few papers have reported on the application of ISA process for the water soluble
PDIs up to now. 16
As discussed above, our studied n-PDI and p-PDI molecules showed a strong
tendency to form aggregates in water due to their strong π-π interaction and hydrophobic
core. Surfactants form micelles above their CMC in water with a hydrophobic core and
hydrophilic corona. Hence, besides the electrostatic interactions between the PDI and
surfactant, the hydrophobic nature of n-PDI or p-PDI (p-PDI is more hydrophobic than n-
73
PDI) will serve as the secondary motifs by interacting with the micelle core to promote
and stabilize the combined surfactant/PDI micelle complex.
3.3.2.1 Photophysics Study upon complexing between PDI and appositively charged
surfactant (n-PDI/DTAC or p-PDI/SDS)
SDS or DTAC solutions were made at different concentrations and then added to
dilute the PDI solutions to vary the PDI/surfactant ratio. The effects were quite dramatic.
Adding DTAC to n-PDI (abbreviated as n-PDI/DTAC) or SDS to p-PDI (p-PDI/SDS)
resulted in an increase in the extinction coefficient, a well-resolved absorption spectra
and a significant increase in the fluorescence intensity above the surfactant CMC (see the
insets in Figure 3.5(a) and (b) and also Table 3.1). n-PDI/DTAC, at the highest DTAC
concentration (35 mM) had an absorption spectrum very similar to n-PDI in DMSO (λ0-0
at 581 nm; ε=3.71 x 104 M-1cm-1 and λ0-1 at 540nm ε=2.28 x 104 M-1cm-1
and a shoulder
at 500nm). Fluorescence emission spectra for n-PDI/SDS showed a blue shift to 606nm
from 616nm in water shown in Figure 3.6b. For n-PDI/DTAC there was an initial
decrease in the fluorescence intensity with the addition of DTAC followed by an
enhancement by a factor of ca. 1.8 above the CMC (quantum yield approximately 0.96).
The strong fluorescence decrease at low DTAC concentration is due to the formation of
non-fluorescent premicellar aggregates.17 It is well known that quaternary ammonium
groups in DTAC could help in stabilizing H-aggregates by forming ion pairs with sulfate
groups.19 This is the highest quantum yield reported for a water soluble PDIs so far.4,6
The blue shift of emission from 616 in water to 606 in the n-PDI/DTAC complex, as
shown in Figure 3.6a, suggests a less polar micelle environment. All results imply that
the n-PDI was monomerically dissolved in the DTAC micelles. It also has a lifetime of
6.91ns, longer compared with n-PDI in water and DMSO indicating the dissociation of n-
74
PDI aggregates when solvated by DTAC micelles. A very high fluorescence rate constant
was obtained for n-PDI in DMSO, compared with its non-radiative rate constant.
For p-PDI/SDS, similar structured absorption spectra were observed (λ0-0 at 592
nm, ε=4.97 x 104 M-1cm-1 and λ0-1 at 547nm, ε=2.71 x 104 M-1cm-1 and a shoulder
around 500nm). The plot of fluorescence intensity versus [SDS] in Figure 3.5b shows a
monotonic increase, ultimately leveling off above 10 mM with an increase by a factor of
ca. 4 (quantum yield about 0.42 after correcting for the OD change at 540nm). The
fluorescence emission spectra showed a blue shift of 14nm relative to that in water, as
shown in Figure 3.6b. The structured absorption spectra and blue shift of the fluorescence
emission spectra indicates the incorporation of single PDI molecules into the SDS micelle
core. Again we observed a double exponential fluorescence decay curve for p-PDI/SDS
with lifetimes of 5.20 and 0.54ns. Very short average lifetime was obtained for p-PDI at
0.72ns. Very high fluorescence rate and non-radiative rate constants are obtained.
3.3.2.2 Photophysics Study upon complexing between PDI and same charged
surfactant (n-PDI/SDS or p-PDI/DTAC)
Even for surfactants with the same charge as the PDI moiety (e.g. p-PDI/DTAC or
n-PDI/SDS), a red shift in absorption spectra and blue shift in fluorescence emission
spectra could also be observed along with an increase in the fluorescence quantum yield
(Table 3.1). The absorption spectra for n-PDI/SDS changed very little with a 12nm red
shift and almost no shift in the fluorescence spectrum compared to water (Figure 3.7a and
b). The fluorescence intensity increased very little over the investigated concentration
range, with a change in the quantum yield from 0.54 in pure water to 0.59 in SDS. A
monoexponential fluorescence decay with a lifetime of 6.09ns was observed. These
results suggest that the incorporation of n-PDI into the SDS micelle is possible but the
75
environment is not appreciably different than in pure water (hence dimers or small
aggregates may be incorporated into the micelles).
In Figure 3.8 is given the absorption and emission spectra of p-PDI upon mixing
with DTAC, which is totally different from the case of n-PDI/SDS. Structured absorption
spectra were observed, similar to p-PDI in DMSO except for the 10nm red shift. The new
peak around 540nm was attributed to the strong hydrophobic interaction between p-PDI
and DTAC core. It is more obvious that p-PDI is incorporated into the DTAC micelle
core based on the 8nm blue shift of emission spectra (Figure 3.8b) and increased
fluorescence quantum yield (Table 3.1). Interestingly, the fluorescence increase in this
case is even higher than that for p-PDI/SDS complexes. The maximum quantum yield
was 0.54, similar to that of p-PDI in DMSO. Again two fluorescence lifetimes were
observed for p-PDI/DTAC, suggesting different local environments for individual p-PDI
moieties.
3.3.2.3 Photophysics Study upon Mixing the PDI/Surfactant complexes
Based on the above experiments we believe that p-PDI and n-PDI were
incorporated into the core of SDS and DTAC micelles, respectively. Therefore,
experiments consisting of mixing p-PDI/SDS and n-PDI/SDS complexes or n-PDI/DTAC
and p-PDI/DTAC complexes were conducted to test this argument. As expected, the PDI
moiety is “protected” by the micelle and there is no fluorescence quenching upon mixing
p-PDI/SDS and n-PDI/SDS. The absorption spectrum upon mixing p-PDI/SDS and nPDI/SDS at 10mM is identical with the average absorption spectrum in Figure 3.9a,
which is also the same result as all other SDS concentrations from 2 to14mM. The
fluorescence intensity as a function of SDS concentration upon mixing the two
complexes agrees very well with the average of fluorescence intensity from the individual
76
complexes. All these results imply that the solubilized p-PDI or n-PDI molecules in
surfactants are confined to the core of SDS or DTAC micelles, and subsequently unable
to complex with each other upon further mixing, unlike the situation in pure water.7
3.3.2.4 Photophysics study of n-PDI titrated by DTAC.
The titration of n-PDI solution by DTAC concentrated solution can provide
further information about the interactions between PDI and surfactant molecules.
Concentrated DTAC solution was prepared in DMF at 0.65M. A very similar absorption
of n-PDI was obtained with the titration of DTAC as for the n-PDI/DTAC complexes
with a new peak emerging at 540nm originating from the interaction between n-PDI and
DTAC molecules (see Figure 3.10a and Figure 3.3a). The fluorescence intensity also
showed a decrease with the initial addition of DTAC and then increased after reaching
the CMC of DTAC (20mM), the same as for n-PDI/DTAC. This indicates that the
presence of n-PDI did not change the CMC of DTAC due to its very low concentration
(about 10-3mM). The quantum yield of n-PDI can reach 95% with [DTAC] ≥ 20mM.
Compared with the 10nm blue shift for n-PDI/DTAC complexes, a blue shift of 12nm
was observed for the n-PDI emission spectra by DTAC, suggesting a less polar
environment the n-PDI molecules.
3.3.3 Photophysics study of n-TDI upon complexing with DTAC surfactant solutions
(n-TDI/DTAC).
Due to very strong π-π interactions, n-TDI is nonfluorescent in water. However,
complexing n-TDI with DTAC surfactant solutions can make n-TDI fluorescent. A
pronounced reduction of intensity of absorption peak near 630nm in favor of band at
longer wavelength of 690nm (λ0,0-1,0) is observed with increasing DTAC concentration
(Figure 3.11a).
The structured absorption spectra indicate the monomeric dispersion of
77
n-TDI into the DTAC micelles. A fluorescence peak at 724nm emerged with a DTAC
concentration higher than 20mM, corresponding to the CMC of DTAC. The fluorescence
intensity saturated after 25mM.
At a concentration of 35mM of DTAC the quantum
yield of n-TDI can reach to 3%. The fast nonraditive rate constant (109s-1) has been
reported for a similar n-TDI compounds which leads to the low quantum yield of n-TDI
in medium.8 Mixing n-TDI/DTAC and p-PDI/DTAC complexes above 20mM of DTAC
showed no fluorescence quenching for either p-PDI or TDI and indicates that the n-TDI
is isolated in the DTAC micelle cores.
3.3.4 Effect of NaCl on the n-PDI and p-PDI Solutions.
The effect on the absorption spectrum of adding NaCl to p-PDI or n-PDI solutions
is shown in Figure 3.12a and b. As can be seen the OD is reduced by about 20% over
the full range of NaCl concentration. We interpret this as a hypochromic effect10 arising
from NaCl-induced further aggregation.
The magnitude of this effect is similar for both
n-PDI and p-PDI but the accompanying decrease in fluorescence is much stronger for pPDI due to the more hydrophobic nature of p-PDI.
The fluorescence “quenching” by
NaCl yields a linear Stern-Volmer curve with slopes 6.23 M-1 and 42.8 M-1 for n-PDI and
p-PDI respectively. As neither the Na+ nor the Cl- ion is expected to quench the
fluorescence by any direct mechanism (e.g. electron transfer or external heavy atom
effects) this result can only be rationalized as the effect of NaCl-induced aggregation. If
we assume that the extent of aggregation is similar for both species (based on the OD
changes) then these results suggest that p-PDI is more prone to self-quenching than nPDI, because of its more hydrophobic nature.6
3.3.5 Effect of PSS-Na+, PDAC+Cl- Polyelectrolytes on n-PDI and p-PDI Solutions.
78
Zimerman et al. have studied the interaction between the water soluble
pyrenetetrasulfonic acid chromophore and cationic polyelectrolytes.19 When the
negatively charged pyrene molecules are mixed with a cationic polyelectrolyte the
excimer fluorescence band of the pyrene is enhanced, presumably because the pyrene
moieties are sequestered within the polyelectrolyte with a high local concentration.
Thus it is reasonable to expect that addition of our charged PDI species to polyelectrolyte
of the opposite charge may result in PDI aggregation.
Xie et al. examined the complex
formed between an unsymmetric cationic PDI species with anionic poly(acrylates) and
found considerable broadening of the absorption and fluorescence spectra, which was
interpreted as the result of PDI excimers (the magnitude of the fluorescence quenching
was not reported).20
Absorption spectra of n-PDI and p-PDI titrated with the oppositely charged
polyelectrolyte are shown in Figure 3.13a and b with the insets showing the fluorescence
quenching curves (log (F0/F) and F0/F respectively). The slow addition of polyelectrolyte
to either PDI solution results in a decrease in the absorbance, accompanied by a slight
change in the spectral shape and fluorescence quenching, especially for n-PDI and
PDAC, as is seen from the inset in Figure 3.13a. A monomer unit concentration of PDAC
of only 2x10-5 M (in polymer repeating units) will decrease the fluorescence by a factor
of ~220 (approximately a 10:1 ratio of PDAC monomer units to n-PDI). The sequestering
of n-PDI by PDAC will lead to an increase of n-PDI’s local concentration and we
presume that the fluorescence quenching is due to the formation of larger aggregates,
even though the spectral changes are relatively minor (but consistent with
hypochromism10). The excited state of n-PDI is expected to be a good electron acceptor
so we would not expect the quaternized amine of PDAC to be an effective fluorescence
79
quencher by an electron transfer mechanism. The quenching of the p-PDI fluorescence
by the slow addition of PSS solution shown in Figure 13b is much smaller compared to nPDI and PDAC, but still substantial, a factor of about 12.
Another set of experiments was carried out in which equal volumes of PDI
solution were diluted all at once with a 1 mg/mL solution of the oppositely charged
polyelectrolyte. Once again a very strong quenching of n-PDI by PDAC was observed
(see Figure 3.13a) with similar spectral changes as illustrated in Figure 3.14a. However,
for p-PDI and PSS the absorption increased and became much more structured (see
Figure 3.14b) and a fluorescence enhancement by a factor of 5 was observed (when
corrected for the increased absorption the fluorescence increase is approximately a factor
of 4), which is similar with that for p-PDI/SDS complexes. For this experiment the ratio
of styrene sulfonate group to p-PDI is approximately 6000:1. If a slowly titrated pPDI/PSS solution is added all at once to excess PSS a fluorescence enhancement is once
again observed, demonstrating that the p-PDI/PSS complex is not kinetically frozen. In
both cases no spectral shift is observed for the fluorescence,.
We assume that all these polyelectrolyte-induced phenomena are related to the
formation of aggregates of the PDI when sequestered within a polyelectrolyte and
furthermore, at least in the case of p-PDI/PSS, the particular nature of the aggregates is
strongly dependent on the p-PDI/PSS ratio. Thus when the PSS is present in large
excess the p-PDI is more isolated from other p-PDI moieties and this is reflected in a
sharpening of the absorption spectrum and an increase in the fluorescence quantum yield.
Thus one can conclude that the p-PDIs form aggregates in pure water, perhaps to a
greater extent than n-PDI, which is why p-PDI has a substantially lower quantum yield
than n-PDI.
When PSS is added slowly at a lower PSS/p-PDI ratio, the local
80
concentration of p-PDI is actually increased, as reflected in the decrease of OD and
fluorescence intensity. The reason PSS is able to break up the p-PDI aggregate is
attributed to the amphiphilic nature of PSS chain. Following the same logic, PDAC
always strongly enhances the aggregation of n-PDI regardless of the n-PDI/PDAC ratio.
3.3.6 Molecular Dynamics Simulation of p-PDI Dimer
Computer simulation of the structure of a possible p-PDI dimer was carried out
using the IMPAC software (with periodic boundary conditions) from the Schroedinger
Corporation.21 For these calculations two p-PDI molecules were placed in the vicinity of
each other with Cl- counterions (for simplicity) and the remaining volume filled with H2O
molecules.
After equilibration of this starting structure the H2O molecules were
removed and then reinserted into the available volume and the structure re-equilibrated
(see Figure 3.15). Then the atomic coordinates from this final structure were used to
calculate the molecular center-to-center distance (7.76 Å) and the value of κ2 = 0.2804,
which is the orientation factor that governs the strength of the dipole-dipole interaction
(see discussion of Förster energy transfer, presented in Chapter 5.2.6). The transition
dipole is assumed to line along the long molecular axis (parallel to the vector joining the
two imide nitrogen atoms).
(µ1· R12)( µ2· R12).
The unit vector along this direction is µi. and κ = µ1· µ2 – 3
From the coordinates we find µ1· µ2 = -0.99792,
(µ1·R12)
=-
0.40077 (µ2· R12) = 0.38961) and κ = 0.2804. An analogous simulation was carried out
for a n-PDI dimer (structure not shown) with a similar center-to-center distance (7.49 Å)
but a much smaller value of κ2 (0.0574).
The center-to-center distances are larger than
one would expect for classical excimer formation, but for this distance and mutual
orientation of the transition dipoles it is plausible that there could be a significant
interaction between the p-PDIs in the excited state, but a much weaker one for the n-PDI
81
dimer. As can be seen from Figure 3.15, the two molecular planes are approximately
parallel (and the twisting of the perylene out of planarity is also easily seen), indicating a
H-aggregate like molecular aggregates.
The structure of the n-PDI dimer is qualitatively
similar.
3.4 Summary
The photophysics of these PDI compounds in water were complicated by the
formation of aggregates, which are evidently fairly weakly interacting (at least with
respect to π−π stacking) given their significant fluorescence quantum yield and the
absence of strongly perturbed absorption or fluorescence spectra. Some aggregates
similar to H-aggregates are formed for p-PDI in water and J-aggregates are formed for nPDI in water. We found that adding classical surfactants (SDS and DTAC) to aqueous
solutions of the PDIs sharpened the absorption spectrum, increased the extinction
coefficient and the fluorescence yield, indicating the breaking up of aggregates of p-PDI
or n-PDI by interacting with surfactant solutions.
We also observed that the extinction
coefficient and fluorescence quantum yield of both n- and p-PDI could be diminished by
the addition of NaCl or low concentrations of polyelectrolytes (PSS or PDAC) to their
aqueous solution.
One would expect NaCl to encourage aggregation by electrostatic
screening and if the PDI compound is sequestered onto a polyelectrolyte of opposite
charge then it is reasonable to think that there could be a high local concentration of the
PDI with concomitant changes in the absorption spectrum and fluorescence quantum
yield.
However mixing of p-PDI with PSS and n-PDI with PDAC at high
polyelectrolyte concentrations provides an interesting contrast.
For p-PDI and a large
excess of PSS the absorption spectrum becomes more structured, the extinction
coefficient increases and the fluorescence yield increases by a factor of approximately
82
three.
Thus in this case we suggest that any p-PDI aggregates in aqueous solution are
broken up and the p-PDI moieties are distributed as isolated monomers along the PSS
chain. For n-PDI and a large excess of PDAC is this not the case and essentially the same
changes in the absorption spectrum and fluorescence yield are observed as for low PDAC
concentrations. We do not have a good explanation for this difference – our LBL
experiments to not indicate that PDAC is an intrinsic quencher for n-PDI (see Chapter 5).
Given this evidence for some kind of “weakly bound aggregate” some molecular
simulations were carried out for two p-PDI molecules interacting in water. According
to these calculations the perylene aromatic rings are approximately co-parallel but offset
from each other with a center-to-center separation on the order of 8 Å, i.e. larger than for
classical excimer formation but close enough for a significant dipole-dipole interaction in
the excited state.
3.5 References
1. (a) Haugland, P. R. Handbook of Fluorescenct Probes and Research Products,
9th ed., Molecular Probes, Eugene, 2002. (b) Seisenberger, G., Ried, M. U.,
Endress, T., Buning, H., Hallek, M., Brächle, C. Science, 2001, 294, 1929. (c)
Cotlet, M., Hofkens, J., Maus, M., Gensch, T., Van der Auweraer, M., Michiels,
J., Dirix, G., Van Guyse, M., Vanderleyden, J., Visser, A. J. W. G., De Schryver,
F. C. J. Phys. Chem. 2001, 105, 4999.
2. Fischer, M., Georges, J. Chem. Phys. Lett. 1996, 260, 115.
3. Margineanu, A., Holfkens, J., Cotlet, M.; Habuchi, S., Stefan, A., Qu, J., Kohl, C.,
, Vercammen, J., Engelborghs, Y., Gensch, T., De Schryver, F. C. J. Phys. Chem.
B. 2004, 108, 12242.
4. Qu, J., Kohl, C., Pottek, M., Müllen, K. Angew. Che. Int. Ed. 2004, 43, 1528.
83
5. (a) Schnupfeil, G., Stark, J., Wöhrle. D. Dyes Pigm. 1995, 27, 339. (b) Langhals,
H., Jona, W., Einsiedl, F., Wohnlich, S. Adv. Mater. 1998, 10, 1022.
6. Kohl, C., Weil, T., Qu, J., Müllen, K. Chem. Eur. J. 2004, 10, 5297.
7. Tang, T. Qu, J. Müllen, K., Webber, S. E. Langmuir 2006, 22, 26.
8. Jung, C., Müller, B., Lamb, D.C., Nolde, F., Müllen, K., Bräuchle, C. J. Am.
Chem. Soc. 2006, 128, 5283.
9. (a) Hofkens, J., Vosch, T., Maus, M., Köhn, F., Cotlet, M., Weil, T., Herrmann,
A., Grebel-Koether, D., Qu, J., Müllen, K., De Schryver, F.C. Chem. Phys. Lett.
2001, 333, 255. (b) Liu, D., De Feyter, S., Cotlet, U.-M., Weil, T., Herrmann, A.,
Grebel-Koether, D., Qu.J., Müllen, K.; De Schryver, F. C. Macromolecules, 2003,
36, 5918.
10. Rhodes, W., Chase, M. Review of Modern Physics 1967, 39, 348.
11. Ford, W. E. J. Photochem. 1987, 37, 189.
12. Würthner, F. Chem. Comm. 2001, 1564.
13. Ford, W. E. J. Photochem. 1987, 37, 189.
14. Faul, C.F.J.; Antonietti, M. Adv. Mater. 2003, 15, 673.
15. (a) Faul, C.F.J.; Antonietti, M. Chem. Eur. J. 2002, 8, 2764. (b) Wei, Z., Laitinen,
T., Smarsly, B., Ikkala, O., Faul, C. F. J. Angew. Chem. Int. Ed. 2005, 44, 751.
16. (a) Franke, D., Vos, M., Antonietti, M., Sommerdijk, N. A. J. M., Faul, C. F. J.
Chem. Mater. 2006, 18, 1839. (b) Marcon, R. O., dos Santos, J. G., Figueiredo, K.
M., Brochsztain, S. Langmuir 2006, 22, 1680.
17. Fendler, J. H. Membrane Minmetic Chemistry John Wiley & Sons, New York,
1982.
84
18. Rosen, M.J. Surfactants and Interfacial Phenomena 2nd ed., Wiley & Sons, New
York, 1989.
19. (a) Zimerman,O.; Cosa, J.J.; Previtail, C.M. J.M.S.-Pure Appl. Chem. 1994, A31,
859. (b) Caruso, F.; Donath, E.; Möhwald, H.; Georgieva, R. Macromolecules,
1998, 31, 7365.
20. Xie, A. F., Liu, B., Hall, J. E., Barron, S. L., Higgins, D.A. Langmuir 2005, 21,
4149.
21. These calculations were carried out by Jessica Kingsberg in the laboratory of Prof.
E. Dormidontova, Department of Macromolecular Science and Engineering, Case
Western Reserve University, Cleveland, Ohio.
85
a
b
O N
O
O N
HO3S
SO3H
O
O
N+
N
O
O
HO3S
O
O
SO3H
O N
O
+
O
O
N+
N+
O
O N
c
O
N
O
NaO3S
SO3Na
O
O
O
O
NaO3S
SO3Na
O
N
O
Figure 3.1 Chemical Structures of n-PDI, p-PDI and n-TDI.
86
O
4MeSO3-
O
N
O
O
N
O
HO3S
O
O
ii:94%
O
O
SO3H
O
O
O
O
HO3S
O
N
HO
O
O
N
SO3H
O
O
N
O
i: 89%
Cl
Cl
Cl
Cl
O
N
n-PDI
HO
O
N
iii: 79%
O
N
O
N
O
N
O
O
O
O
N
O
N
O
O
N+
O
O
iv:94%
N
N
N+
O
O
4MeSO3-
N+
N+
O
N
O
p-PDI
Figure 3.2 Synthetic Scheme for n-PDI and p-PDI. i) Phenol, N-methyl-2
pyrrolidinone(NMP), K2CO3, 80ºC, 15hrs, 89%. ii) Conc. Sulfuric acid r.t.
15hrs, 93%. iii) 4-[2-(dimethylamino)ethyl]phenol, NMP, K2CO3, 80ºC,
15hrs, 79%. iv) silver methanesulfonate, r.t., 5hrs, 95%.
87
a
1
n-PDI
Fluo. Inten.(a.u.)
Absorbance
0.06
0.8
0.04
0.6
0.4
0.02
0.2
0
0
b
0.06
Absorbance
400
460
520
580
640
Wavelength(nm)
700
760
1
p-PDI
0.8
0.04
0.6
0.4
0.02
Fluo. Inten.(a.u)
340
0.2
0
0
340
Figure 3.3
400
460
520
580
640
Wavelength(nm)
700
760
(a) and (b) Absorption and emission spectra of 1.5 x 10-6 M n-PDI and pPDI in water (solid line) and DMSO (dotted line), respectively.
88
UV n-PDI
UV p-PDI
UV p-PDI/n-PDI
UV p-PDI+n-PDI
FL n-PDI
FL p-PDI
FL p-PDI/n-PDI
FL p-PDI+n-PDI
Absorbance
0.08
0.06
0.04
0.02
Fluo.Inten.(a.u.)
0.1
0
340
Figure 3.4
400
460
520 580 640 700
Wavelength(nm)
760
820
880
Absorption (left) and fluorescence emission (right) spectra in water for nPDI (black line), p-PDI (dark blue line), p-PDI/n-PDI mixed in a 1:1 ratio,
(red line) and the sum of the p-PDI and n-PDI spectra (light blue line).
89
2
a
0.06
Fluo. Inten.
Absorbance
0.08
n-PDI
1.5
1
0.5
0
0.04
0
5
10
15 20 25
[DTAC]/mM
30
35
40
0.02
0
400
500
600
700
Wavelength(nm)
900
6
0.08
Fluo. Inten.
b
0.06
Absorbance
800
4
2
0
0.04
0
0.02
2
4
6
8
[SDS]/mM
10
12
14
p-PDI
0
400
500
600
700
Wavelength(nm)
800
900
Figure 3.5 Absorption spectra of n-PDI with DTAC (a) and p-PDI with SDS (b) at
different concentrations. Insets show the fluorescence intensity as a function
of surfactant concentration. Arrow indicates increasing surfactant
concentration.
90
a
n-PDI
5mMDTAC
10mMDTAC
15mMDTAC
20mMDTAC
25mMDTAC
30mMDTAC
35mMDTAC
Fluo. Inten.
n-PDI/DTAC
560
600
640
680
Wavelength(nm)
720
760
800
p-PDI
b
Fluo. inten.
p-PDI/SDS
2mMSDS
4mMSDS
6mMSDS
8mMSDS
10mMSDS
12mMSDS
14mMSDS
560
600
640
680
720
Wavelength(nm)
760
800
Figure 3.6 Fluorescence emission spectra of n-PDI (a) and p-PDI (b) mixing with SDS
and DTAC surfactant solutions at different concentrations, respectively.
91
0.1
a
Absorbance
0.08
0.06
0.04
0.02
0
300
400
500
600
Wavelength(nm)
700
800
1.4
Fluo. Inten.
Fluo. Inten.
b
1.2
1
0.8
0
560
600
2
640
680
Wavelength(nm)
4
720
6
8
10
[SDS]/mM
760
12
14
800
Figure 3.7 (a) and (b) Absorption and emission spectra of n-PDI upon mixing with SDS
solutions at different concentrations. Ex=540nm. Inset in (a) showing the
fluorescence intensity (corrected to the abs) as a function of SDS
concentrations. Absorption and emission spectra of n-PDI in water are
shown as the red line. The arrow indicates the concentration increasing of
SDS.
92
0.06
p-PDI
a
0.05
5mMDTAC
Absorbance
10mMDTAC
0.04
15mMDTAC
20mMDTAC
0.03
25mMDTAC
30mMDTAC
0.02
35mMDTAC
0.01
0
340
420
500
580
660
Wavelength(nm)
740
820
900
Fluo. Inten.
6
Fluo. Inten.
d
4
2
0
0
560
600
5
640
680
Wavelength(nm)
10
15
20
[DTAC ]/mM
720
25
760
30
35
800
Figure 3.8 (a) and (b) Absorption and emission spectra of p-PDI upon mixing with
DTAC solutions at different concentrations. Ex=540nm. Inset in (b)
showing the fluorescence intensity as a function of DTAC concentration.
The arrow indicates the concentration increasing of DTAC.
93
a
0.08
Fluo. Inten.
Absorbance
0.1
0.06
0.04
0.02
0
360
420
480
540
600
660
Wavelength(nm)
720
780
Fluo. Inten.
b
2
Figure 3.9
4
6
8
10
[SDS]/M
12
14
16
(a) Absorption and emission spectra of p-PDI/SDS mixing with nPDI/SDS at [SDS]=10mM (Red) and the average absorption spectrum of
individual p-PDI/SDS and n-PDI/SDS at [SDS]=10mM(black). (b)
Fluorescence intensity upon mixing p-PDI/SDS and n-PDI/SDS (red circle)
and averaging the individual p-PDI/SDS and n-PDI/SDS (black triangle) as
a function of SDS concentration.
94
0.03
Absorbance
a
0.02
0.01
0
340
400
460
520
580
640
Wavelength(nm)
700
760
Fluo. Inten.
b
0
560
600
640
5
10
15
20
25
DTAC Concentration/mM
680
720
Wavelength(nm)
760
30
800
Figure 3.10 (a) and (b) Absorption and emission spectra of n-PDI titrated by
concentrated DTAC DMF solution. Red curves represent the spectra of nPDI in water. Inset shows the fluorescence versus DTAC concentration and
the arrow indicates increasing DTAC concentration.
95
0.06
Absorbance
a
0.04
0.02
0
400
460
520
580
640
700
760
Wavelength(nm)
1
b
Fluo. Inten.
Fluo. Inten.
0.8
0.6
0.4
0.2
0
0
660
700
740
5
780
820
Wavelength(nm)
10
15
20
25
DTAC Concentration/mM
860
900
30
35
940
Figure 3.11 (a) and (b) Absorption and emission spectra of n-TDI upon mixing with
DTAC solutions at different concentrations. Ex=640nm. Inset in (a) showing
the fluorescence intensity as a function of DTAC concentrations. Absorption
and emission spectra of n-TDI in water are shown as the red line. The arrow
indicates the concentration increasing of DTAC.
96
3
a
2
NaCl
Increases
0.06
F0/F
Absorbance
0.08
2.5
1.5
1
0.5
0.04
0
0
0.04
0.08
0.12
0.16
[NaCl](M)
0.2
0.24
0.02
0
400
460
520
580 640
700
Wavelength(nm)
0.12
820
880
7
0.08
6
5
NaCl
increases
0.06
F0/F
b
0.1
Absorbance
760
4
3
2
1
0
0
0.04
0.02
0.04
0.06 0.08
[NaCl]/M
0.1
0.12
0.02
0
400
460
520
580 640 700 760
Wavelength(nm)
820
880
Figure 3.12(a) Absorption spectra of n-PDI and (b) p-PDI as a function of NaCl
concentration, with insets showing the S-V fluorescence quenching curves.
The arrow indicates increasing NaCl concentration.
97
a
Absorbance
0.1
PDAC
Increase
0.08
Log[F0/F]d
0.12
2
1
0
0.06
0.0E+00
0.04
1.0E-05 2.0E-05 3.0E-05
monomer Comcentration/M
4.0E-05
0.02
0
Absorbance
0.08
460
520
760
820
880
10
b
PSS
increases
0.06
580
640
700
Wavelength(nm)
0.04
F0/F
400
5
0
0.0E+00
3.0E-06 6.0E-06 9.0E-06
Monomer Concentration(M)
0.02
0
400
500
600
700
Wavelength(nm)
800
900
Figure 3.13 Absorption spectra for n-PDI (a) and p-PDI (b) with the addition of PDAC
and PSS solutions with insets showing the fluorescence ratio log(F0/F) or
F0/F, respectively. Arrow indicates increasing PDAC and PSS
concentration. Red curve represents the spectrum in water.
98
0.08
a
n-PDI
Fluo. Inten.
Absorbance
0.06
0.04
0.02
0
300
400
500
600
700
800
Wavelength(nm)
0.08
b
p-PDI
Fluo. Inten.
Absorbance
0.06
0.04
0.02
0
400
450
500
550
600
650
Wavelength(nm)
700
750
800
Figure 3.14 (a) Absorption and fluorescence spectra of n-PDI before (black) and after
(red) mixing with 1 mg/ml PDAC solution, demonstrating fluorescence
quenching under these conditions (b) Absorption and fluorescence spectra of
p-PDI before (black) and after (red) mixing with 1 mg/ml PSS solution,
demonstrating fluorescence enhancement under these conditions.
99
Figure 3.15 MD structure for p-PDI dimer in pure water
100
Table 3.1 Photophysical parameters for n-PDI and p-PDI in different solutions
Media
n-PDI
p-PDI
Water
DMSO
DTAC
SDS
Water
DMSO
DTAC
SDS
λ0,0-0,1
(nm)
568
577
581
580
587
581
593
592
λ1,0-0,0
(nm)
618
611
606
617
628
612
622
614
εmax
(M-1cm-1)
29700
34400
37100
30200
32800
38500
62900
49700
ΦFa
τb
(ns)
4.78
5.49
6.91
6.03
6.35(0.52); 0.13(0.48) [3.36]c
5.41
5.61(0.75); 0.78(0.25) [4.40]c
5.20(0.04); 0.54(0.96) [0.72]c
0.54
0.82
0.96
0.59
0.14
0.55
0.52
0.42
kf
(x10-8s-1)
1.13
1.49
1.39
0.98
0.26
1.02
1.18
5.83
knr
( x10-8s-1)
0.96
0.33
0.059
0.68
1.59
0.83
1.09
8.06
.
a. The quantum yield (ΦF) was measured relative to the standard PDI compound N,N’-bis(2,6-dimethylphenyl)-3,4,9,10perylenetetracarboxylic diimide in dichloroethane, for which ΦF = 1 is assumed.
b.Values in ( ) are the pre-exponential factors in a bi-exponential fit.
c. The average fluorescence lifetime by weighing the pre-exponential factors.
101
Chapter 4 Molecular Layer-by-Layer Self-assembly of Water Soluble
Dyes through the π-π and Electrostatic Interactions
4.1 Introduction
Conventionally, the Layer-by-Layer (LBL) process is based on the sequential
adsorption of polycations and polyanions from dilute aqueous solution onto a solid
support with film formation a consequence of the electrostatic interaction and complex
formation between the oppositely charged polyelectrolytes.1 Functional films for making
thin-film filed-effect transistiors and photovoltaics require the preparation of well defined
films composed of molecules with appropriate properties in a specified geometrical
arrangement with respect to each other and the substrate. Research on zwitterionic
molecules assembled alternately has been carried out to achieve these goals although the
integrity of the films was not easily controlled.1a The interpenetration of polyelectrolyte
chains between adjacent layers has been addressed.2 Therefore, precise sterochemical and
positional control of the functional species at the molecular level have not been achieved
by the LBL self assembly technique to our knowledge. The Langmuir-Blodgett (LB)
deposition technique offers control over the architecture of the films at the molecular
level to offer certain desirable properties tailored to suit the application required.
However, it is very difficult for the technique and complicated for the instrument and the
molecules suitable for this technique are very limited.3 On the other hand, the further
application of conjugate polymers or organic molecular semiconductors as the organic
based FET or OLED has been limited due to their low electron mobility between the
crystal “grains” boundaries. The π stacking is expected to provide more efficient orbital
overlap and thereby facilitate the carrier transport.4 In this Chapter we will discuss the
102
fabrication of self-assembled thin films of perylene diimides carrying four positive or
negative charges respectively (p-PDI and n-PDI see Figure 3.1) on quartz or silicon
substrates using molecular layer-by-layer (MLBL) deposition from aqueous solution. The
application of this technique to some other charged aromatic species (pyrenetetrasulfonic
acid tetrasodium salt (n-PSA) and terylene diimide(n-TDI)) was also investigated. The
reasons for choosing PDIs as the building block are: 1) isolated PDI moieties have a large
extinction coefficient in the visible, a high fluorescent quantum yield and very good
photostability and are good candidates for n-type semiconductors, photon harvesting
arrays (photovoltaic cells) and OLEDs; 2) the perylene structure is very liable to form
assemblies because of the strong π-π interaction between them and the π stacking is
expected to facilitate energy or charge transport.4,5 The π stacking can also lead to
strong quenching of the fluorescence, as we will see in the following.
Thus the
fluorescence intensity serves as a measure of the extent to which the perylene diimide
moieties can be isolated from each other.
4.2 Experimental Section
4.2.1 Materials
The p-PDI (CAS[817207-11-7]), n-PDI (CAS[694438-88-5]) and n-TDI, whose
chemical structures shown in Figure 3.1, were synthesized by Müllen et al.8
Pyrenetetrasulfonic acid tetrasodium salt (n-PSA) was purchased from Molecular Probes
and used as received. Molecular dynamics simulations indicated that p-PDI has the
dimensions of 2.13x1.33x1.89 nm (LWH) and n-PDI 2.13x1.15x1.49 nm (LWH) (Figure
3.1). The perylene core itself has a dimension of 0.72x0.53 nm (LW). The peak molar
extinction coefficients in water were measured with 3.28x104 and 2.97x104 M-1 cm-1 for
103
p-PDI (λ = 591 nm) and n-PDI (λ = 568 nm) respectively. Beer’s law was verified for
these solutions, suggesting that aggregation of the two PDIs is absent over this
concentration range. As discussed in Chapter 3, both of these compounds are fluorescent
in the aqueous phase, with the n-PDI having a higher quantum yield of fluorescence by a
factor of approximately 4 (φfln-PDI = 0.54 and φflp-PDI = 0.14). However if n-PDI and pPDI are mixed in a 1:1 ratio the fluorescence is strongly quenched (by a factor of ca. 13
relative to p-PDI at an equivalent optical density). As will be discussed later, an
analogous quenching also occurs in the MLBL assembly. The absorption spectra of nTDI was shown in Figure 3.11 with it chemical structure illustrated in Figure 3.1. The
molar extinction coefficients of n-TDI are 4,000(437nm), 29,000(640nm) and 18,000 M1
cm-1 (685nm). Chemical structure of n-PSA is illustrated in Figure 4.5.
4.2.2 Molecular Layer-by-Layer procedures.
The MLBL procedure was carried out by dipping the cleaned quartz substrates
alternately into the 0.l g/l p-PDI or n-PDI (or n-TDI or n-PSA) solutions for 10 minutes,
starting with the p-PDI layer. After each dipping, there were three washes with DI water
to remove the physically absorbed PDI dye. After this step, a gentle stream of argon gas
was used to dry the films before they were put into the next negatively charged dye
solution. All the absorption and fluorescence spectra were taken after each dipping and
drying process. These routines were repeated until desired number of double layers was
achieved. The films are denoted in the following by (p-PDI/n-PDI or n-PSA or n-TDI)n
where n is the number of double layers.
4.2.3 Instruments and measurements
104
UV-Vis and fluorescence spectra were collected on the same instruments
described in chapter 2 and chapter 3. For the AFM measurements a commercial
microscope (D-3000, Digital Instruments, Santa Barbara, CA) was used. Images were
recorded in tapping mode in air employing cantilevers with a nominal radius of less than
10 nm, resonance frequency at 300 kHz and a spring force constant at 40N/m.
4.3 Results and Discussions
4.3.1 Molecular Layer-by-Layer of p-PDI and n-PDI
4.3.1.1 Absorption spectra of (p-PDI/n-PDI)n
The growth of the MLBL PDI thin films was examined by using UV-Vis
absorption spectroscopy. A typical series of absorption spectra for the formation of (pPDI/n-PDI)n films on the quartz is shown in Figure 4.1a and Figure 4.1b shows the
absorbance at 577nm as a function of the number of monolayers of p-PDI and n-PDI. The
strong absorbance between 350nm and 650nm is the typical absorption band of the
perylene diimides.
For the first two layers there is a small (5nm) blue shift in the peak
positions (Table 4.1). After the deposition of several p-PDI/n-PDI double layers, the
maximum of the absorption peak for perylene diimide core shifted from 567nm to
577nm. The direct exposure of the first two PDI layers to the quartz surface might
contribute to the slight blue shift. A very systematic change in the absorption intensity for
both perylene diimide cores and phenyl groups can be easily observed with the increasing
of PDI layers, which shows the consistency of the amount of absorbed PDI molecules in
the film. The absorbance for phenyl groups as a function of number of PDI layers is
shown in Figure 4.1c. The absence of any obvious change such as broadening or shift of
the absorption spectra implies that no further aggregates were formed upon MLBL
105
process. For PDI compounds without the bay substituents there is a very strong red-shift
in the
absorption and fluorescence spectrum upon aggregation.6
The good linearity of
the absorbance with the number of PDI monolayers indicates a constant increase in the
density of PDI molecules deposited into the film upon each dipping cycle, even for the
first one or two monolayers. The peak positions for UV-Vis absorption of (p-PDI/n-PDI)n
are the same as those for the mixing of p-PDI and n-PD((1:1 ratio) in solution shown in
Figure 3.4, which indicates that the incorporated p-PDI and n-PDI is in a 1:1 ratio for
each bilayer.
Control experiments dipping only into a p-PDI solution failed to increase the
absorbance progressively, even after 10 dipping cycles, which implies that the π-π
interaction alone was not sufficient to make the PDI layers (see Figure 4.2).
The following rough calculation suggests that only a monolayer of p-PDI or nPDI was formed for every deposition. The surface density, Γ, of PDI for every deposition
in (p-PDI/n-PDI)n films can be calculated using Γ =[Aλελ-1NA]x10-3, where Aλ is the
absorbance of PDI in the film at a given wavelength, ελ is the extinction coefficient of
PDI in solution (M-1 cm-1) at λ, and NA is Avogadro’s number.7 Using the average
molar extinction coefficient of 3.0x104 M-1 cm-1 at 578nm for p-PDI and n-PDI, an
average surface PDI surface density of 3.64x1013 molecules per cm2/layer is obtained for
(p-PDI/n-PDI)n films, which corresponds to an average area per PDI of 2.78 nm2.
Assuming a single PDI molecule occupies an area of approximately 2.62 (LxW) nm2 (the
average from the earlier estimate of molecular dimensions) the average surface coverage
is ~94% which reflects close packing of PDI molecules in each monolayer.
4.3.1.2 Characterization of (p-PDI/n-PDI)n
106
The TM-AFM image shown in Figure 4.3b yields a surface roughness of 2.4nm
for (p-PDI/n-PDI)10 annealed at 80ºC over night. As we can be seen from the image,
some domain structures exist, which can be attributed to aggregates of either p-PDI or nPDI in water. Measuring the depth of a scratch made on this sample indicates a total film
thickness of 14 nm, leading to a PDI layer thickness of 0.70nm. This value is smaller than
the PDI thickness (dimension H above) obtained from the molecular dynamic simulation,
suggesting the collapse of the PDI aggregates upon deposition. Ellipsomerty was also
used for the measurements of the film thickness. For this experiment (p-PDI/n-PDI)n was
built onto a silicon wafer. For the fitting of the experiment data, a bare silicon wafer was
used to build up a SiO2/Si model and then it was used for the fitting of the experiment
data over 700nm for the thickness, because there is no absorbance for the PDI films
above 700nm. All the MSE values after fitting were less than 10. The thickness curve is
shown in Figure 4.3b. A thickness of 0.56nm for each PDI monolayer can be obtained by
this approach. The refractive index was fitted to be around 1.57.
4.3.1.3 Emission spectra of (p-PDI/n-PDI)n
Figure 4.4 displays the fluorescence spectra of (p-PDI/n-PDI)n films (excitation
wavelength at 540nm) and the fluorescence intensity as a function of number of PDI
monolayers. For all the fluorescence spectra only one peak can be observed and the shape
and wavelength maxima are very similar to the solution phase spectra.6 The peaks are
blue shifted for the first two PDI monolayers and then reach the same peak position at
614nm, which is very consistent with the UV-Vis spectra. Note that the fluorescence
intensity decreased after the first double layer, indicating the presence of strong
fluorescence quenching with the absorption of second PDI double layer. After that, the
107
fluorescence increased linearly with the number of PDI monolayers. The intensity when
n-PDI is on the outside layer is higher than for p-PDI as the outside layer due to the
higher quantum yield of n-PDI. We also note that there is essentially no shift in the
fluorescence maximum compared to the solution phase. The fact that the fluorescence
intensity is sensitive to the outermost layer suggests that excitation energy is transferred
to the outer layer before fluorescence occurs (e.g. we hypothesize that the outer layer is
acting as a energy trap, regardless of which PDI species is present). If there were no
energy transfer, then the fluorescence intensity should be just a simple average of the
contribution from each species, since our film is optically thin (e.g. all chromophores
have an equal probability to be excited).
By carefully comparing the fluorescence intensity of the films with an aqueous
solution of p-PDI contained in a small path length cell with a similar OD (the thin cell
was used to eliminate any optical artifacts in the front-face fluorescence mode) we find
that the fluorescence intensity for the (p-PDI/n-PDI)n film is approximately 13 times
lower than for the solution species, just as was found for the 1:1 complexes formed in the
solution phase(see Chapter 3).
4.3.2 Molecular Layer-by-Layer of p-PDI and n-PSA
Instead of using two very similar molecules for MLBL, the smaller dye molecule
PSA was also investigated to test its feasibility for MLBL. The photophysics of n-PSA in
water is shown in Figure 4.5a along with the chemical structure of n-PSA. A very
structured absorption and emission spectra were observed with three absorption peaks at
379nm, 354nm, and 335 nm (attributed to the transition of S0-S1 for 0,0-1,0, 0,0-1,1, 0,0-
108
1,2, respectively). n-PSA is very fluorescent in water with a QY close to unity and a.
Stokes shift of about 21nm.
4.3.2.1 Absorption spectra of (p-PDI/n-PSA)n
The absorption spectra of (p-PDI/n-PSA)n were shown in Figure 4.5b and the
absorbance at 379nm for PSA (corrected for a slight contribution from p-PDI) and
575nm for p-PDI is plotted as a function of number of double layers in Figure 4.5c. The
absorption for p-PDI lies primarily in the range of 400nm to 600nm and n-PSA 340nm to
400nm. The n-PSA absorption is considerably less structured than observed in the
solution phase.
No formation of n-PSA aggregates could be observed as seen from
Figure 4.5a. No obvious stripping was found for either p-PDI or n-PSA deposition. The
absorbance for p-PDI and n-PSA increases linearly with the number of p-PDI layers.
However, the loading efficiency for p-PDI is not as high as for the p-PDI/n-PDI case. The
uptake of p-PDI for (p-PDI/n-PSA)n is about 70% of that for (p-PDI/n-PDI)n.
4.3.2.1 Emission spectra of (p-PDI/n-PSA)n
The emission spectra of (p-PDI/n-PSA)n films were taken using excitation
wavelengths at both 360nm and 540nm. The emission spectra for the 1st layer of p-PDI
and (p-PDI/n-PSA)1 are illustrated in Figure 4.6a. As we can see there is fluorescence
emitted from the 1st layer of p-PDI due to the weak absorbance of p-PDI at 360nm. The
fluorescence centered at 610nm is from S1-S0 and 420nm from S2-S0. The fluorescence
for S2-S0 is due to the twist of the perylene diimide as had been reported by Qu et al..8
Upon deposition of n-PSA onto the p-PDI layer, no fluorescence from n-PSA could be
observed but there was a fluorescence increase for p-PDI. We attribute this to the energy
transfer from n-PSA to p-PDI. The subtraction of the 1st layer fluorescence of p-PDI from
109
the fluorescence of (p-PDI/n-PSA)1 is shown as the black curve in Figure 4.6a, indicating
strong energy transfer from n-PSA to p-PDI. If the (p-PDI/n-PSA)n-1/p-PDI films were
excited at 540nm, their emission spectra are given in Figure 4.6b. For n=1, 2, and 3, the
fluorescence was strongly quenched and then changed little after that. We attribute this
quenching effect to the electron transfer mechanism within the p-PDI and n-PSA layer.
This scheme is shown in Figure 4.6c. After photoexcitation, the exciton was formed in
p-PDI layer. Because n-PSA has larger band gap, one electron from HOMO of n-PSA
can be transferred to the hole of HOMO in p-PDI*, leading to the fluorescence quenching
of p-PDI. The very close packing of the double layers makes this process possible (the
thickness of each layer in (p-PDI/n-PDI)n is 0.6nm). While with increasing of double
layers (n), the ratio of number of layers for p-PDI and PSA became closer to 1, the
fluorescence became little changed.
4.3.3 Molecular Layer-by-Layer of (p-PDI/n-TDI)n
A larger water soluble π-conjugated molecule, n-TDI, was also investigated for its
suitability for the MLBL process. Its absorption and fluorescence spectra were shown in
Figure 3.11 in chapter 3. It is believed that H-aggregates of n-TDI are formed in water,
due to its non-fluorescent behavior in water.9
4.3.3.1 Absorption spectra of (p-PDI/n-TDI)n
The absorption spectra for (p-PDI/n-TDI)n were demonstrated in Figure 4.7a. The
spectra showed a very systematic increase of both p-PDI and n-TDI. No stripping effect
was observed for either p-PDI or n-TDI deposition, indicating the strong interaction
between n-TDI and p-PDI layers. The deposition of n-TDI will lead a little increase in the
p-PDI absorption region due to the spectral overlap between p-PDI and n-TDI. The
110
absorbance from p-PDI and n-TDI increased linearly with the number of double layers as
shown in Figure 4.7b. The absorbance from p-PDI was corrected from n-TDI absorbance
at 575nm. It is worth mentioning that the loading efficiency of p-PDI is higher than that
for (p-PDI/n-PDI)n. This may be due to the larger aggregates formed by n-TDI in water.
A molecule similar to n-TDI was shown to have the aggregates on the order of 100nm in
water.9
The absorbance at 284nm for phenyl groups on both p-PDI and n-TDI increased
linearly with the number of deposition monolayers, illustrated in Figure 4.7c.
4.3.3.2 Emission spectra of (p-PDI/n-TDI)n
Emission spectra from (p-PDI/n-TDI)n were shown in Figure 4.8. The emission
spectrum from the 1st p-PDI layer was shown as the red curve in Figure 4.8 and then the
fluorescence was totally quenched upon the next deposition of n-TDI layer. No
fluorescence emission was observed for the following p-PDI and n-TDI depositions. Due
to the large spectra overlap between the p-PDI emission spectra and n-TDI absorption
(see later), the energy transfer should be very efficient between the p-PDI and n-TDI
layers.
4.3.4 Molecular Layer-by-Layer of (p-PDI/(n-TDI:n-PSA))n
A mixture solution (0.1g/l) of n-TDI and n-PSA (1:1 by mol:mol) was employed
for the further investigation of molecular dimension on the behavior of MLBL. The
absorption spectra for (p-PDI/(n-TDI:n-PSA))n is shown in Figure 4.9a. Very similar
spectra to (p-PDI/n-TDI)n were observed, except that there was also an absorbance from
n-PSA around 360nm. The p-PDI loading these two cases is about the same (compare
Figure 4.9b and 4.7b). However, The TDI uptake for each deposition in the n-TDI: nPSA mixture is about 75% of that in n-TDI solutions alone (comparing Figure 4.9b and
111
4.7b). The smaller uptake of n-TDI was compensated by the incorporation of n-PSA into
the films to finish the charge neutralization and over compensation. Again, no n-PSA
extraction was observed upon p-PDI deposition. This experiment demonstrated that the
larger π-conjugate molecule (n-TDI) is preferentially absorbed onto the p-PDI layer.
4.3.5 Energy Transfer for TDI/(p-PDI/n-PDI)n
FRET has been frequently employed as a useful “spectroscopic ruler” for the
studying the macromolecular interaction in biological or polymeric systems.10 In this
study it is utilized to study the PDI organization and proved the layered PDI structure in
the formed films of (p-PDI/n-PDI)n. n-TDI was selected as the acceptor and absorbed
onto the quartz as the foundation layer. Films of (p-PDI/n-PDI)n-1/p-PDI were built up on
the n-TDI foundation layer. The strong spectral overlap between fluorescence of the (pPDI/n-PDI)10 MLBL and the absorption of n-TDI
is shown in Figure 4.10, which can
lead to long-range energy transfer from (p-PDI/n-PDI)n(Donor) to the n-TDI(Acceptor)
layer. The R0 for (p-PDI/n-PDI)n films with n-TDI can be calculated according to the
following equations.
8.75 × 10 −5 κ 2QD J
R0 =
n4
(1)
J = ∫ FD ( λ )ε A (λ )λ4 dλ
(2)
0
Where κ2 is the orientation factor between donor and acceptor molecules, QD0 is
the quantum yield of donor fluorescence in the absence of acceptor, and n is the index of
refraction(here is taken as 1.57 as obtained from the ellipsometry measurements). J is the
spectral overlap between the donor emission spectrum and acceptor absorption spectrum.
FD is the normalized donor fluorescence spectrum and εA is the wavelength dependent
112
molar extinction coefficient of the acceptor. R0 is in units of angstroms, whereas λ is in
nanometers and εA in M-1cm-1. J was obtained at 2.78994×1015, leading to a R0 of 5.74nm
if κ2 is assumed to be 4 (PDI and n-TDI dipole moments are parallel).
4.3.5.1 Absorption spectra for n-TDI/(p-PDI/n-PDI)n
The absorption spectra of n-TDI/(p-PDI/n-PDI)n were shown in Figure 4.11a,
indicating the progressive building up of the p-PDI/n-PDI MLBL films on top of TDI
monolayer. Absorption spectra of n-TDI/(p-PDI/n-PDI)n after subtracting the absorption
spectrum of n-TDI foundation layer are shown in Figure 4.11b. The smaller peak around
700nm is interpreted as arising from the interaction between the n-TDI and deposited (pPDI/n-PDI)n. This peak was also observed in the case of (p-PDI/n-TDI)n. In Figure 4.11c
is plotted the absorbance at 580nm and 700nm as a function of the number of PDI layers.
The PDI loading for each layer deposition is about the same as that for (p-PDI/n-PDI)n
discussed earlier as illustrated in Figure 4.1.
4.3.5.2 Energy Transfer for n-TDI/(p-PDI/n-PDI)n
The dependence of the donor and acceptor separation distance (d) on the donor
fluorescence intensity can be written as: 2b
Id
1
=
(3)
I 0 1 + (d 0 / d ) x
Where Id is the fluorescence intensity for donor molecules spaced an average
distance of d away from the center of acceptor, I0 is the fluorescence intensity of the
donor molecules in the absence of acceptor molecules, and d0 is the critical spacing
between donor and acceptor corresponding to a 50% probability of energy transfer. Both
d0 and x are a function of the geometry assumed. For example, x is equal to 6 in case of a
donor molecule transferring energy to an acceptor molecule in 3-D space, x is equal to 4
113
in the case of a plane donor molecules to a plane of acceptor molecules, and x is equal to
2 in case of energy transfer from a J-type aggregates in a planar array.
Eq. 4 can be obtained by the transforming of Eq. 3, shown as the following:
Log[(I0/Id)-1]=-x Log(d)+ x Log(d0)
(4)
Therefore, the plotting of Log[(I0/Id)-1] versus Log(d) can yield the two
parameters x and d0.
Since the molecular layer-by-layer technique is used, we first assumed that these
MLBL films are comprised of discrete, non-interpenetrating layers of PDI monolayer. As
discussed before, each monolayer of PDI is about 6Aº thick (dPDI). The estimation of the
average distance d from (p-PDI/n-PDI)4/p-PDI to n-TDI is illustrated in Figure 4.12a.
The thickness of the n-TDI monolayer (dTDI) is assumed to be 12Aº.
d for nth p-PDI layer can be derived as follows:
d=1/2dTDI+1/2(2n-1)dPDI=6n+3 Aº
(5)
The fluorescence intensity of n-TDI/(p-PDI/n-PDI)n-1/p-PDI and (p-PDI/n-PDI)n1/p-PDI
versus the number of p-PDI layers(n) are shown in Figure 4.12b. For nth p-PDI
layer, I0 is the fluorescence from (p-PDI/n-PDI)n-1/p-PDI and Id is from
n-TDI/(p-
PDI/n-PDI)n-1/p-PDI. A significant fluorescence quenching was observed during the first
several p-PDI layers. After 7 p-PDI layers, the fluorescence increased linearly with the
number of p-PDI layers, but still with a decrease in fluorescence intensity compared with
the MLBL films for p-PDI/n-PDI without a n-TDI foundation layer.
The plot of Log[(I0/Id)-1] vs. Log(d) is shown in Figure 4.12c, yielding as the
slope x = 2.03 with an intercept of 3.61. It has been reported that x =2 is appropriate for
energy transfer from J-aggregates monolayer assemblies to an acceptor monolayer.10 This
114
indicates that the orientation of PDI within the MLBL films is analogous to J-type
aggregates. The value of d0 was obtained from the intercept is 60.4Aº. The energy transfer
efficiency for n-TDI/(p-PDI/n-PDI)n-1/p-PDI is shown in Figure 4.13. The 50% energy
transfer corresponds to 9.5 p-PDI layers, resulting in a d0 of 60Aº from Eq. 5, which is
very consistent with the calculated value of 57.4Aº and simulated value of 60.4Aº.
Another approach was used to model energy transfer in the p-PDI/n-PDI films. In
this model, we will deal with the energy transfer for the nth p-PDI as the average
contribution from 1st, 2nd,·····, n-1th, nth layer of p-PDI. Therefore, the donor fluorescence
dependence can be formulated in the following equation:
In 1 n
1
= ∑
I 0 n 1 1 + (d 0 / d n ) x
(6)
Where d0 is assumed to be equal to 60Aº(what we got from above discussion) and
the calculation of dn is shown in Eq.5. The plot of In/I0 as a function of p-PDI layers for
the experimental data and simulated data for x=2, 3 and 4 is shown in Figure 4.13b. The
best fit with experimental data is in case that x is equal to 2, which is consistent with the
conclusion we obtained above.
4.4 Summary
A new layer-by-layer technique that uses a combination of π-π and electrostatic
interactions has been shown to be very versatile for making monolayer self-assembled
films from the water soluble π-conjugated dye molecules, which we refer to as the as
Molecular Layer-by-Layer (MLBL in short).
We have demonstrated that it is possible to fabricate monolayer assemblies by the
MLBL technique for two very similar dye molecules such as p-PDI and n-PDI. Both the
115
absorption and emission spectra of MLBL films are identical to the complex of p-PDI: nPDI (1:1) in aqueous solutions. The surface coverage of PDI in the monolayer was
calculated to be as high as 94%. However, smaller domains were observed from the AFM
image with a RMS roughness of around 2.4nm. This domain structure was attributed to
the PDI aggregates present in aqueous solutions, as illustrated in Chapter 3. The strongly
quenched fluorescence is evidence for π-stacking and the dependence of the fluorescence
intensity on the outermost layer suggests efficient energy transfer between the PDI
moieties.
In addition to using very similar size dye molecules, a smaller dye molecule ( nPSA) and larger molecule (n-TDI) were also investigated for the fabrication of MLBL
films with p-PDI. Comparing with (p-PDI/n-PDI)n, a smaller loading efficiency was
observed for (p-PDI/n-PSA)n and a higher loading efficiency for (p-PDI/n-TDI)n was
observed. Very efficient energy transfer was observed from n-PSA to p-PDI upon the
excitation of n-PSA. When p-PDI was excited at 540nm, fluorescence quenching was
also observed, which we presume originates from electron transfer from n-PSA to p-PDI*
(Figure 4.6c). If a 1:1 mole:mole mixture of n-TDI and n-PSA was utilized as the dipping
solution for the negatively charged layer, MLBL films were built up for (p-PDI/(n-TDI:nPSA))n. The p-PDI uptake under this condition is very similar with the case for (p-PDI/nTDI)n. However, this experiment demonstrated that the larger π-conjugate molecule (nTDI) is preferentially absorbed onto the p-PDI layer, almost certainly a reflection of the
strong π-stacking interaction. The uptake of n-TDI is about 75% of that for (p-PDI/nTDI)n. No extraction of dye molecules was observed upon depositing of the next dye
molecule layer.
116
An energy transfer study for n-TDI/(p-PDI/n-PDI)n indicated that the PDI
orientation in the planar films is analogous to J-type aggregation. The critical energy
transfer distance was estimated from calculation and experiment to be approximately
60Aº.
We think the MLBL technique could be applied for making films for future
applications in thin film transistors, OLED or other organic-photonic devices from
appropriate charged conjugated dye molecules.
4.5 References
1. (a) Decher, G., Hong, J. Makromol. Chem. Symp. 1991, 46, 321. (b) Decher, G.
Science 1997, 277, 1232.
2. (a) Schlenoff, J. B., Dubas, Stephan, T., Farhat, T. Langmuir 2000, 16, 9968. (b)
Baur, J. W., Rubner, M. F.; Reynolds, J. R.; Kim, S. Langmuir 1999, 15, 6460.
3. Ulman,A. An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to
Self Assembly. Academic Press: New York, 1991, Chapter 3.
4. (a) Moon, H., Zeis, R., Borhent, E.J., Besnard, C., Lovinger, A. J.; Siegrist,T.;
Kloc, C.; Bao, Z. N. J. Am. Chem. Soc. 2004, 126, 15322. (b) Curtis, M.D., Cao,
J., Kampf, J.W. J. Am. Chem. Soc. 2004, 126, 4318.
5. (a) Malenfant, P.R.L., Dimitrakopouls, C.D., Gelmore, J.D., Kosbar, L.L.,
Graham, T.O., Curioni, A., Andreoni, W. Appl. Phys. Lett. 2002, 80, 2517. (b)
Chesterfield, R.J.; Mckeen, J.C.; Newman, P.R.; Ewbank, P.C.; da Silva Filho, D.
A., Brédas, J-L., Miller, L. L.; Mann, K. P., Frisbie, C.D. J. Phys. Chem. B, 2004,
108, 19281.
117
6. (a) Law, K.Y. Chem. Rev. 1993, 93, 449. (b) Liu, S-G., Sui, G., Cormier, R.
A., Leblanc, R. M., Gregg, B. A. J. Phys. Chem. B 2002, 106, 1307. (c) Wang,
W., Han, J. J., Wang,L-Q., Li, L-S., Shaw; W. J., Li, A. D. Q. Nano Lett. 2003, 3,
455.
7. (a) Caruso, F., Kurth, D. G., Volkmer, D., Koop, M. J., Müller, A. Langmuir
1998, 14, 3462. (b) Li. D., Swanson, B. I., Robison, J. M., Hoffbauer, M. A.
J.
Am. Chem. Soc. 1993, 115, 6975.
8. Qu. J., Zhang, J., Drimsdale, A. C., Müllen, K., Jaiser, F., Yang, X., Neher, D.
Macromolecules 2004, 37, 8197.
9. Jung, C., Müller, B., Lamb, D.C., Nolde, F., Müllen, K., Bräuchle, C. J. Am.
Chem. Soc. 2006, 128, 5283.
10. (a) Stryer, L. Annu. Rev. Biochem. 1978, 47, 819. (b) Bokinsky, G.; Zhuang,
X.W. Acc. Chem. Res. 2005, 38, 566.
11. Nakahara, H., Fukuda, K., Möbius, D., Kuhn, H. J. Phys. Chem. 1986, 90, 6144.
12. Johnson, G. E. Macromolecules 1980, 13, 145.
118
0.06
Absorbance
a
0.04
0.02
0
Absorbance at 580nm
250
0.04
350
450
550
Wavelength(nm)
650
750
850
b
0.03
0.02
0.01
0
0
2
4
6
8
10
12
14
Number of PDI Layers
4
6
8
10
12
Number of PDI Layers
16
18
20
0.06
c
Absorbance at 284nm
0.05
0.04
0.03
0.02
0.01
0
0
2
14
16
18
20
Figure 4.1 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and n-PDI
(Red). (b) The peak absorbance around 575 nm as a function of monolayer PDI with pPDI as the first layer. (c) The peak absorbance around 284nm as a function of monolayer
PDI with p-PDI as the first layer.
119
Absorbance
0.006
0.004
0.002
0
250
350
450
550
Wavelength(nm)
650
750
Figure 4.2 The absorption spectra of p-PDI dipping cycles up to 10. Each dipping is
about 10 minutes.
120
30
b
Thickness(nm)
25
20
y = 0.5699x
15
10
5
0
0
10
20
30
40
50
Number of PDI Layer
Figure 4.3 (a) The TM-AFM image of (p-PDI/n-PDI)10 annealed at 80ºC overnight. (b)
Ellipsometric film thickness as a function of PDI monolayers.
121
Fluo. Inten.
a
560
610
660
710
Wavelength(nm)
760
Fluo. Inten.(a.u.)
b
0
1
2
3
4
5
6
7
8
Number of Double Layer
9
10
11
Figure 4.4 (a) Fluorescence emission spectra from the monolayer of p-PDI/n-PDI with pPDI as the first layer (b) The fluorescence intensity as a function of double layer p-PDI/nPDI (diamonds correspond to n-PDI as the outside layer and the circles p-PDI as the
outside layer). The lines between points illustrate the order of deposition (e.g. p-PDI
first, followed by n-PDI).
122
a
Absorbance
0.04
NaO4S
SO4Na
NaO4S
SO4Na
0.03
0.02
Fluo. Inten.
0.05
0.01
0
300
350
Absorbance
0.016
400
Wavelength(nm)
450
500
b
0.012
0.008
0.004
0
300
400
500
600
Wavelength(nm)
700
800
Absorbance
0.016
c
0.012
0.008
0.004
0
0
1
2
3
4
5
6
7
Number of Double layer
8
9
10
Figure 4.5 (a) Absorption and emission spectra of n-PSA in water with the inset showing
n-PSA’s chemical structure. Ex=360nm. (b) Absorption spectra of monolayer of p-PDI
and n-PSA starting with p-PDI layer. (c) Peak absorbance as a function of double layer
(p-PDI/n-PSA) for p-PDI at 575nm (circles) and n-PSA at 379nm (diamonds) after
correcting from the absorbance at 379nm for p-PDI.
123
Fluo. Inten.
a
380
420
460
500
540
580
Wavelength(nm)
620
660
700
Norm. Fluo. Inten.
1
Fluo. Inten.
b
0.8
0.6
0.4
0.2
1
560
610
c
2
3
4
5
6
7
8
9
Number of p-PDI Layers(n)
660
710
Wavelength(nm)
10
760
LUMO
LUMO
eT
HOMO
HOMO
n-PSA
p-PDI*
n-PSA
p-PDI
Figure 4.6 (a) Emission spectra of 1st p-PDI (Red) monolayer and (p-PDI/n-PSA)1
(Pink). Black curve is the subtraction of 1st p-PDI and (p-PDI/n-PSA)1. Ex=360nm. (b)
Emission spectra of (p-PDI/n-PSA)n-1/p-PDI with Ex=540nm. (c) Electron transfer
mechanism from the HOMO of PSA to p-PDI exciton.
124
0.06
a
Absorbance
0.05
0.04
0.03
0.02
0.01
0
240 300 360 420 480 540 600 660 720 780 840 900
Wavelength(nm)
0.03
Absorbance
b
0.02
0.01
0
Abosrbance at 284nm
0
1
0.05
2
3
4
5
6
Number of Double Layers
7
8
9
c
0.04
0.03
0.02
0.01
0
0
1
2
3
4 5 6 7 8 9 10 11 12 13 14 15
Number of Depsition Layers
Figure 4.7 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and n-TDI
(Red). (b) The peak absorbance around 575 nm for p-PDI (black circles) and 700nm for
n-TDI (red diamonds) as a function of double layers. (c) The peak absorbance around
284nm as a function of deposition layers with p-PDI as the first layer.
125
Fluo. Inten. (a.u.)
560
600
640
680
Wavelength(nm)
720
760
Figure 4.8 Emission spectra of (p-PDI/n-TDI)n with Ex=540nm.
126
800
0.05
a
Absorbance
0.04
0.03
0.02
0.01
0
240 300 360 420 480 540 600 660 720 780 840 900
Wavelength(nm)
0.03
Absorbance
b
0.02
0.01
0
0
1
2
3
4
5
6
Number of Double layers
7
8
9
Figure 4.9 (a) UV-Vis absorption spectra of the monolayer of p-PDI (Black) and nTDI:n-PSA (Red). (b) The peak absorbance around 575 nm for p-PDI (black circles) and
700nm for n-TDI (red diamonds) as a function of the number of double layers
127
Fluo. Intensity or ε
560
610
660
710
760
Wavelength(nm)
810
860
Figure 4.10 Emission spectrum of (p-PDI/n-PDI)10(Red) and absorption spectrum of nTDI (Green).
128
Absorbance
0.06
a
0.04
0.02
0
Absorbance
340
420
500
580
660
Wavelength(nm)
740
820
900
b
0.06
0.04
0.02
0
c
0.06
0.04
0.0021
0.02
0
Absorbance at 700nm
Absorbance at 579nm
250 300 350 400 450 500 550 600 650 700 750 800 850
Wavelength(nm)
0.0042
0
0
2
4
6 8 10 12 14 16 18 20 22 24
Number of PDI Deposition Layers
Figure 4.11 (a) UV-Vis absorption spectra of n-TDI/(p-PDI/n-PDI)n. Green: n-TDI,
Black: p-PDI and Red: n-PDI. (b) Absorption spectra of n-TDI/(p-PDI/n-PDI)n upon
subtracting the absorption spectrum from n-TDI foundation layer. Black: p-PDI and Red:
n-PDI. (c) The peak absorbance around 579 nm as a function of monolayer PDI with pPDI as the first layer. The absorbance was corrected from n-TDI spectrum (black circles).
Red diamonds represent the absorbance around 700nm shown in Figure 4.11b as a
function of the number of PDI deposition layers.
129
a
d
Fluo.Inten.
b
0
1
Log[(I0/Id)-1]
2.5
2
3
4
5
6
7
8
Number of p-PDI layers
9
10
11
12
c
2
1.5
1
0.5
y = -2.0269x + 3.6097
0
-0.5
0.7
0.9
1.1
1.3
1.5
Log(d)
1.7
1.9
2.1
Figure 4.12 (a) Schematic representation of deriving the average distance from (p-PDI/nPDI)4/p-PDI to the center of n-TDI layer. (b) Fluorescence emission intensity for (pPDI/n-PDI)n and n-TDI/(p-PDI/n-PDI)n as a function of PDI double layers. (c) Plot of
Log[(I0/Id)-1] versus Log(d), yielding a slope of 2.02 and intercept 3.61.
130
Energy Transfer Efficiency
1
0.8
a
0.6
0.4
0.2
0
1
0.6
3
4
5
6
7
8
Number of p-PDI Layers
9
10
11
12
b
0.5
I0/Id
2
0.4
0.3
Exp. Data
x=2
x=4
x=6
0.2
0.1
0
0
1
2
3
4
5
6
7
8
Number of p-PDI Layers
9
10
11
12
Figure 4.13 (a) Energy transfer efficiency as a function of number of p-PDI layers. (b) pPDI fluorescence dependence as a function of number of p-PDI layers. ● represents the
experimental data; ▲: x=2; ♦: x=4; ■: x=6.
131
Chapter 5: Incorporation of Water Soluble Perylene Diimides into
Polyelectrolyte Films by the Layer-by-Layer Method
5.1 Introduction
Organized perylene diimide thin films have been prepared previously by vapor
deposition, spin casting, self-assembly via hydrogen bonding and electrostatic interaction,
and Langmuir-Blodgett techniques.1 These films have been shown potential applications
for making heterojunction photovoltaics and n-type organic semiconductors.
The discovery of layer-by-layer (LBL) technology has inspired researchers to
make functional films with desired compositions and controlled thickness during the last
15 years.2 Because of its simplicity, reproducibility and film uniformity over a large area
the LBL technique has been widely used in fabricating all kinds of hetero-structural films
incorporating dendrimers, quantum dots, polymer micelles, proteins and even carbon
nanotubes in support of applications such as chemical sensors, fuel cells, nano- or ultra
filtration, OLED etc.3
Recently the introduction of small dye molecules into the LBL film fabrication
has drawn considerable interest due to potential applications such as photon harvesting
arrays, OLED, controlled drug release and so on. However, a typical problem
encountered during the LBL fabrication process is the extraction of small molecules by
the next polyelectrolyte layer deposition.4 In this chapter we demonstrate that water
soluble PDIs can avoid this problem. The Förster Resonance Energy Transfer study
between PDI layers and brilliant green was carried out to elucidate the film structure. In
possible future work the incorporation of PDI into films composed of water soluble
conducing polymers such as sulfonate polyaniline will be very interesting for the study of
132
OLED and photovoltaics and photon harvesting studies instead of the inert
polyelectrolytes such as PSS or PDAC discussed in this Chapter.
5.2 Experimental Section
5.2.1 Materials
Same as stated in the experimental sections in Chapter 3.2.1 and Chapter 4.2.1.
5.2.2 Layer-by-Layer Procedures
The quartz substrates were dipped in fresh Piranha solution for 10 minutes
(H2SO4:H2O2=7:3 v/v, extremely corrosive, be careful!!) followed by RCA solution for
10 minutes (NH4OH:H2O2:H20=6:2:2, v/v/v), both at 60°C, and then baked at 420 ºC for
6 hrs. The advancing contact angle with water was measured to be approximately 5º for
all the substrates. Then the quartz was coated with the PEI layer to make it evenly
charged with a high density of cations. The LBL dipping time for all solutions was 10
minutes, with both sides of the quartz substrate exposed to the dipping solution. A gentle
stream of argon gas was employed to dry the films between each dipping step. For n-PDI
LBL film fabrication a double layer of PSS/PDAC was used as a foundation layer before
the n-PDI layer was deposited. Films with 1, 3 and 5 separation layers (SLs) between the
adjacent n-PDI layers were fabricated, whose film structure can be denoted as
PEI/PSS/PDAC/(n-PDI/PDAC/(PSS/PDAC)m)n. 1, 3, and 5 SLs correspond to m = 0, 1
and 2 respectively. If p-PDI was used as the PDI building block, the films can be
represented as PEI/PSS/PDAC/PSS/(p-PDI/PSS/(PDAC/PSS)m)n with a 3 layer
foundation. It is worth mentioning that the number of PSS layers in p-PDI LBL films is
nearly twice that in n-PDI films (for n-PDI films the number of PSS layers is NPSS = 1 +
mn and for p-PDI films NPSS =
2 + (m + 1)n). Films with 2SLs between p-PDI and n-
133
PDI layers were also fabricated, whose film structure can be symbolized as
PEI/PSS/PDAC/PSS/(p-PDI/PSS/PDAC/n-PDI)n. “Precursor films” of PEI/(PSS/PDAC)n
(n=1, 2 and 3) were fabricated in 0.5 M NaCl for the study of the rate of diffusion of nPDI into the LBL films.
For the study of the rate of diffusion of p-PDI into LBL films
precursor films of PEI/(PSS/PDAC)n/PSS (n = 0, 1, and 2) were fabricated in 0.5 M
NaCl.
5.2.3 Spin-Assisted Layer-by-Layer (SA-LBL)
Brilliant Green (BG) was used as energy acceptor for the energy transfer study.
However, standard dipping experiments showed significant BG penetration into the LBL
films, which totally quenched the PDI fluorescence. Therefore, a strategy of spin-casting
BG on top of the dried non-swollen solid films was employed so that the short exposure
time of the BG to the LBL would reduce the BG penetration. BG was immediately spun
cast at a speed of 2000 rpm after being drop cast onto the PDI films and then the film was
washed four times by DI water at the same speed. BG was spun-coated onto both sides
of the LBL films.
5.2.4 UV-Vis and Fluorescence Spectroscopy Measurements
Both UV-Vis and fluorescence spectroscopy were measured to monitor the LBL
deposition process. Absorption spectra were obtained with a HP 8453 diode array
spectrometer. The fluorimeter used was a SPEX Fluorolog-τ2 equipped with a 450 W
xenon
light
source,
Czerny-Turner
double
grating
excitation
and
emission
monochromators. A photomultiplier voltage of 950V was typically used and the
excitation and emission slit widths were set at 2/2/4/4mm for LBL samples. The quartz
disk was fixed into a homemade sample holder that allowed the fluorescence to be
134
collected in front face mode. In order to average out positional uncertainties all the
fluorescence spectra were obtained in triplicate for every PDI layer deposition.
5.2.5 Atomic Force Microscopy (AFM) Measurements
For the AFM measurements a commercial microscope (D-3000, Digital
Instruments, Santa Barbara, CA) was used. Images were recorded in tapping mode in air
employing commercial available cantilevers with a nominal radius of less than 10 nm,
resonance frequency at 300 kHz and a spring force constant at 40N/m. The rms surface
roughness of n-PDI LBL films were 3.5nm and 2.6 nm, before or after n-PDI deposition.
5.2.6 Förster Resonance Energy Transfer (FRET) Study
FRET has been frequently employed as a useful “spectroscopic ruler” for the
studying the macromolecular interaction in biological or polymeric systems.5 In this
study it is utilized to study the PDI distribution in the fabricated LBL films. R0, which is
the characteristic distance at which 50% energy transfer efficiency is achieved, is
calculated by the following equations:
6
R0 =
9000(ln 10)κ 2 * QD
J
128π 4 Nn 4
∞
J =∫
0
FD (ν )ε A (ν )
ν4
dν
(1)
(2)
3
 7.66 × 10−8 

C0 = 
(3)
R0


QD is the quantum yield of the donor (p-PDI or n-PDI); FD(ν) is the normalized
fluorescence spectrum of the donor; εA(ν) is the molar extinction coefficient of the
acceptor in M-1cm-1, J is overlap integral with the units cm4M-1cm-1, which is further
expressed in Eq. 2, and ν is the wave number in cm-1.5 <κ2> is the orientational factor
135
for dipoles which is taken to be 2/3 for the R0 calculations presented here. In
homogeneous solution the characteristic concentration (C0) required for 76% quenching
is given by Eq. 3, given in mM in the following. The absorption and emission spectra
for p-PDI or n-PDI overlap, and their R0 values for self-transfer in aqueous solutions
were calculated to be 3.58 nm (C0 = 9.79 mM) and 4.50 nm (C0 = 4.93 mM),
respectively. R0 values for p-PDI/BG and n-PDI/BG are 4.7 and 5.2 nm, corresponding to
C0 at 4.3 and 3.2mM, respectively. The rate of energy transfer is also depends on the
orientation of the transition dipoles of the donor and acceptor and is usually expressed by
the equation
kET = kD0 <κ2> (R0/RDA)6
(4)
In this equation <κ2> reflects the appropriate average value of the angular
dependence of the dipole-dipole interaction and can vary from 0 to 4.
For rapidly
rotating molecules the value is 2/3, as we used for the above calculations, but for
randomly oriented and static molecules the value of <κ2> is 0.476.5b All J, R0 and C0 are
listed in Table 5.1. These R0 values are on the order of the thickness of individual
deposited polyelectrolyte layers and suggest that inter- and intra-layer energy transfer
could occur (see later discussion).
5.3 Results and Discussions
5.3.1 Dye Penetration into PEI/(PSS/PDAC)n LBL Films
A control experiment was carried out to assess the ability of n- or p-PDI to diffuse
into “precursor films” PEI/(PSS/PDAC)n (n = 1, 2 and 3 were investigated here) (see
Experimental). The absorbance and fluorescence intensity as a function of the dipping
time is illustrated in Figure 5.1a and b respectively. The absorption and fluorescence
136
emission spectra for n = 3 are similar to the spectra of the PEI/PSS/PDAC/(nPDI/PDAC/PSS/PDAC)n LBL film (see Figure 5.2-5.3). The time required for the n-PDI
concentration to reach equilibrium in the film is about 10 minutes, which sets the
disposition time scale for the PDI LBL film fabrication. It can be seen that for thicker
PEI/(PSS/PDAC)n films more n-PDI is absorbed. This implies the presence of extensive
penetration and migration of the PDI molecules or their solution-phase aggregates within
the LBL films, such that the PDI is not necessarily deposited in a well-defined “layer”.
Similar experimental results were obtained for p-PDI and (PSS/PDAC)n/PSS (n=0, 1 and
2) LBL films, except that the fluorescence intensity decreased for longer dipping times.
This indicates that self-quenching of p-PDI occurred after being incorporated into the
polyelectrolyte. The absorbance and fluorescence intensity as a function of dipping time
were illustrated in Figure 5.3.
5.3.2 LBL of n-PDI or p-PDI with PSS and PDAC
5.3.2.1 LBL with 0.5M NaCl in Polyelectrolyte Solution
A
typical
series
PDI/PDAC/PSS/PDAC)n
of
absorption
spectra
of
PEI/PSS/PDAC/(n-
with 3 SLs and different numbers of n-PDI deposition layers
are shown in Figure 5.4a, together with the inset showing the absorbance at the maximum
(~580 nm) as a function of number of n-PDI layers. In this case, PSS and PDAC
solutions were made in 0.5M NaCl. The absorption spectrum is almost identical to the
complex between n-PDI and PDAC in homogeneous solution (Figure 3.13a). The
maxima peak position for n-PDI LBL with 3SL is located at 576nm, which is an 8 nm red
shift compared to n-PDI in aqueous solution. The linearly increased absorbance of the nPDI is unchanged when the next double layer (PDAC/PSS) is deposited (see the inset of
137
Figure 5.4a), which indicates that no n-PDI was extracted by the subsequent PDAC and
PSS depositions. Figure 5.4b shows the same result when p-PDI is used as the building
block with PSS and PDAC for making PEI/PSS/PDAC/PSS/(p-PDI/PSS/PDAC/PSS)n.
The absorption is identical to the complex between p-PDI and PSS in homogeneous
solution. Again, no extraction of p-PDI was observed upon the deposition of PSS/PDAC
double layer (see inset in Figure 5.4b. This behavior is unusual as extraction of small dye
molecules is often observed by the deposition of subsequent polyelectrolyte layers.4
The emission peak position for the fluorescence in 3SL n-PDI LBL films is at 614
nm (Figure 5.5a), a 3 nm blue shift comparing with the n-PDI solution.6 The fluorescence
intensity showed a good linearity with the number of n-PDI layers, although there is a
discernable change in the fluorescence intensity when the polyelectrolyte double layer is
deposited on top of the PDI. The non-zero intercept of the fluorescence intensity vs. the
number of n-PDI layers indicates that the fluorescence quantum yield is diminished for n
> 1, presumably because of additional self-quenching and interlayer FRET. A typical
double layer of PSS/PDAC has a thickness about 12nm, as verified by the ellipsometry
measurements.7 The R0 for n-PDI/n-PDI is 4.5nm. Therefore, no efficient FRET will
occur if the n-PDI is sharply confined to a single thin layer. The dipping experiments
with precursor films showed the strong tendency of PDI to penetrate into the LBL films.
Therefore, if there is migration of n-PDI within the films this may permit FRET.
By
carefully comparing the fluorescence intensity of the films with an aqueous solution of nPDI contained in a small path length cell with a similar OD (the thin cell was used to
eliminate any optical artifacts in the front-face fluorescence mode) we find that the
fluorescence intensity for the LBL film is approximately 33 times lower than for the n-
138
PDI solution species. This value is much smaller than the quenching factor (~220) of nPDI by PDAC in homogeneous solution (see Figure 3.13a and 3.14a, Chapter 3). We
ascribe this loss of fluorescence to self-quenching as the local concentration of the PDI
moieties is estimated to be fairly high (see later discussion).
The fluorescence spectra of p-PDI LBL films having 3 SLs are displayed in
Figure 5.5b along with the inset showing the fluorescence intensity versus the number of
p-PDI layers. A single emission peak was observed at 620nm, with a red shift about 2 nm
compared to the p-PDI solution. A quenching factor of 25 was obtained from the thin cell
measurements, compared to an enhancement by a factor of ca. 4 for the p-PDI/PSS
complexation at higher PSS concentration. We always find that the deposition of PSS
onto p-PDI increases the fluorescence intensity and the subsequent deposition of PDAC
decreases it. This is illustrated in Figure 5.6a for p-PDI and 5 SLs. We can propose
that the first increase with the addition PSS layer is due to the migration of p-PDI
molecules into it, thereby decreasing of the p-PDI local concentration and hence
diminishing self-quenching (a large excess of PSS can also increase p-PDI’s fluorescence
intensity in the solution phase, as shown in Figure 3.14b). Addition of the PDAC layer
reverses this result as the PDAC competes with the p-PDI for the anionic PSS groups,
once again compressing the p-PDI layer and enhancing self-quenching. If this is the
mechanism for the phenomenon presented in Figure 5.6a, then it surprises us that it
persists for so many depositions and implies considerable co-mixing of the individual
layers in the LBL film.31 In Figure 5.6b is given the n-PDI LBL films with 5SLs. It can
be seen that the fluorescence did not change as much as that for the p-PDI case (Figure
5.6a). The deposition of PDAC following the n-PDI layers leads to a smaller fluorescence
139
decrease, but not as much as that for n-PDI quenched by PDAC in homogeneous solution
(Figure 3.13a and Figure 3.14a). Analogous to the p-PDI results above, depositing PSS
next to PDAC resulted in an appreciable fluorescence increase.
5.3.2.2 LBL with no NaCl in Polyelectrolyte Solution
In Figure 5.7a and b are displayed the UV-Vis absorption spectra of n-PDI or pPDI with PSS and PDAC with 3 SLs without the use of NaCl during the LBL procedure.
Insets showed the absorbance around 580nm as a function of PDI layers. Similar spectra
and peak positions were observed as when 0.5 M NaCl was used. However, two main
different features were observed here compared with the LBL when 0.5 M NaCl salt was
employed. Firstly, the loading capacity of PDI into the films is about 4x smaller than
when NaCl is present. We presume that this phenomenon originates from the fact that
thicker deposition layers of PSS and PDAC are achieved when 0.5 M NaCl was used
during the LBL process.7 Secondly, some minor extraction of n-PDI by PDAC or p-PDI
by PSS was observed under this deposition condition. This is undoubtedly a result of the
diminished screening which enhances the electrostatic attraction between the PDI and the
oppositely charged polyelectrolyte. The enhanced extraction of n-PDI by PDAC/PSS
compared with that of p-PDI by PSS/PDAC is due to the less hydrophobic nature of nPDI.
Figure 5.8a and b display the fluorescence emission spectra as a function of the
number of n-PDI and p-PDI layers respectively with 3SLs under the condition where no
NaCl was used. The insets showed the fluorescence intensity as a function of PDI layers.
Peak positions were centered at 614nm and 625nm for n-PDI and p-PDI LBL, which are
corresponding to the blue shift of 4nm and 5nm comparing to their aqueous solution
140
phases, respectively. It can see clearly seen that the overall fluorescence intensities for the
same PDI layer with no NaCl used during the LBL procedures are higher than that with
0.5 M NaCl utilized (Figure 5.5a and b correspondingly). Thin cell experiments showed
quenching factors of about 15 and 18 for n-PDI and p-PDI LBLs relative to their solution
phases. This is attributed to the fact that the loading efficiency of PDI is lower when no
NaCl was used in the dipping media, leading to less self-quenching. An obvious
fluorescence increase was observed after carrying out a double layer polyelectrolyte
deposition on top of n-PDI. As has been shown in Figure 5.5b, this might be due to the
compressing of n-PDI from PDAC layer to n-PDI layer. Therefore, the quenching from
PDAC is diminished.
Formation of LBL films of p-PDI and n-PDI with 1 and 5 SLs was also carried
out. The very strong interactions between PDIs can make it possible to build up the
PDI/PSS or PDAC LBL films with 1SL. The characteristics of p-PDI and n-PDI LBL
films such as maxima peak positions for both absorption and fluorescence emission
spectra, slopes of absorbance as a function of the number of PDI layers and quenching
factor (e.g. comparison with the thin cell measurements) are given in Table 5.2 for 1, 3,
and 5 SLs. As we can see in Table 5.2, the amount of PDI deposited (abs slope) was
essentially independent of the number of SLs (1, 3 and 5), implying that each PDI layer
(which presumably achieved charge overcompensation) acts as a barrier for the next one
deposited. When 0.5M NaCl was used during the LBL, the loading efficiency of PDI is
about 4 times higher than that when no NaCl was used. The fluorescence quenching is
diminished as the number of SLs increases, presumably because there is less selfquenching due to the migration of PDI into the adjacent polyelectrolyte layers. This effect
141
is also illustrated in Figure 5.9, which shows the fluorescence intensity increasing
steadily with the number of separation layers but the amount of PDI increases linearly
with the number of PDI depositions. Therefore, the separation layers between the PDI
layers behave like inert buffer layers that will accommodate the same amount of PDI
molecules but with a lower overall local concentration.
5.3.2.3 Characterization of p-PDI or n-PDI in the LBL films
Two TM-AFM images are illustrated in Figure 5.10a and b for p-PDI LBL films
with 3SLs and 0.5 NaCl used for the polyelectrolyte solutions. The surface became
smoother when p-PDI was absorbed on top of the films. The RMSs before and after pPDI deposition were measured to be 3.5 and 2.6 nm, respectively, which was similar to
the case of 1, 3, 6, 8-pyrene tetrasulfonic acid salt self assembled with PSS and PAH.4b
In the literature, two main species of PDI aggregates are distinguished: Haggregates with the face-to-face or card-pack stacking configuration leading to an
absorption blue shift in respect to the solution phase and fluorescence annihilation and Jaggregates with the head-to-tail configuration leading to the red shift of both absorption
and fluorescence spectra.8 For our PDI LBL films there is only a slight red shift of the
absorption spectra and blue shift of the emission spectra, compared with their solution
phases. Thus we conclude that there is no evidence for the formation of H- or Jaggregates in our fabricated LBL films, and therefore the n-PDI or p-PDI moieties are
randomly embedded into the adjacent polyelectrolyte films (due to the fact that the
loading efficiency for 1SL is the same as that for 3 and 5 SLs). However, as we will see
later, there are very strong interactions between the loaded PDIs.
142
Based on the estimated polyelectrolyte
layer thickness of 6 nm when 0.5M NaCl
is used,7 the local n-PDI and p-PDI concentration in a single layer can be deduced from
Beer’s Law as 0.64M and 0.56 M respectively (this estimate ignores diffusion of the PDI
into adjacent layers, which as discussed above almost certainly occurs). Therefore, the
loading efficiency of PDI in the LBL films is much higher than that for the NaCl-free
polyelectrolyte solutions. This concentration corresponds to a volume per molecule of ca.
2.3-2.6 nm3, which is smaller than the approximate molecular volume in water (3.65 nm3)
estimated from MD simulations (see 3.9). Thus there are ample opportunities for
molecular interactions or the formation of larger aggregates that could lead to quenching.
These estimated local concentrations are also much higher than C0 (see eq 3 above) for
these molecules, suggesting that facile intralayer Förster energy transfer could occur. If
there is no NaCl utilized during the LBL process, the local concentration of n-PDI and pPDI was calculated on the order of 0.25M. As discussed above, the decrease of the
quenching factor with the number of SLs is attributed to the n-PDI or p-PDI’s partial
migration into the adjacent layers. The diffusion away from such a high local
concentration of PDI into the adjacent layers could easily provide a concentration higher
than C0 (9.79 and 4.93mM for p-PDI and n-PDI respectively Table 5.1), thus leading to
the interlayer FRET. The scheme in Figure 5.11 represents these ideas.
5.3.3 Energy Transfer Study Between PDI films and BG.
The following experiments were designed to test the interlayer energy transfer
efficiency, since other workers have suggested LBL films containing dyes as photon
harvesting devices.9 Our objective in these experiments was to place a layer of an energy
acceptor at the outermost layer of a film containing PDI molecules and see if there was
143
extensive energy migration between the PDI layers by measuring the fluorescence
quantum yield as a function of the number of films cast. Brilliant Green (BG), which
has been widely used as a optical probe or biological dye was chosen as a energy
acceptor due to its good spectral overlap with the PDI emission (see earlier R0 and C0
estimates, Table 5.1)
As was pointed out above, the double layer thickness of
PDAC/PSS is around 12 nm, 7 in which case FRET between the PDI moieties in different
layers is not expected to be efficient (see the R0 estimates in Table 5.1). Since we found
that BG from bulk water is able to penetrate into the as-formed LBL films, a spin-casting
technique was employed in an attempt to confine the BG to the outermost layer of the
LBL film (see the Experimental Section), but as we will see in the following, it does not
seem that confinement of the BG to the outermost layer was very successful.
The FRET efficiency was measured by comparing the fluorescence of a particular
film before and after deposition of BG. First we measured the FRET efficiency versus
the concentration of spin-casting BG solution for the film PEI/PSS/PDAC/(nPDI/PDAC/PSS/PDAC)3/n-PDI/PDAC/PSS/BG (2SL between outermost n-PDI and BG
layers,
see
Figure
5.12a)
and
for
the
film
PEI/PSS/PDAC/PSS/(p-
PDI/PSS/PDAC/PSS)3/p-PDI/PSS/PDAC/PSS/BG (3SLs between outermost p-PDI and
BG layers see Figure 5.12b). Both curves are very similar except that the BG
concentration for n-PDI films is much larger than for p-PDI to achieve similar quenching
efficiencies. Likewise the plot of FRET vs. the number of layers is very similar for n-PDI
and p-PDI except for the very large difference in the BG concentration required (Figure
5.13a and b). Given the similarity in the R0 values for n-PDI and p-PDI for energy
transfer to BG the data in Figure 5.13 cannot be due to the long distance energy transfer
144
between PDI molecules but instead indicate extensive BG penetration into the films (at
least on the order of 6-7 layers). We suggest that the very large difference between the
required BG concentrations to achieve efficient PDI quenching is because of the different
local environment of the PDI moiety. A layer of p-PDI is surrounded on both sides by
PSS, which would be expected to strongly attract the cationic BG. Thus if there is
significant intermingling of the PSS layers and the p-PDI, the BG could be brought into
relatively close proximity to the p-PDI.10 By the same argument, the BG would be
excluded from the immediate vicinity of the n-PDI (these qualitative ideas are
represented in Figure 5.14). Thus it would seem that these experiments designed to test
for inter-layer energy transfer have revealed more about the diffusion of a relatively large
dye molecule (BG) into the film and the segregation of this dye within the multilayer
assembly.
5.4 Summary
In this work we have explored the incorporation of charged water soluble perylene
diimide dyes into polyelectrolyte multilayers using the layer-by-layer (LBL) dipping
technique. An underlying motivation for this study is the potential for such layers to act
as photon harvesting arrays given the fairly large R0 values for same-species Förster
resonance energy transfer (calculated to be 3.58 nm and 4.50 nm for p-PDI/ p-PDI and nPDI/ n-PDI respectively), which are on the order of the thickness of individual
polyelectrolyte layers. These PDI species have relatively high fluorescence quantum
yields in water and it was hoped that this property would be retained in polyelectrolyte
films, but in fact there was quite considerable fluorescence quenching (by factors from 9
145
to 43 – see Table 5.2), which we have interpreted as self-quenching (we estimate the local
concentration in the layers to be on the order of 0.5 M).
Both p-PDI and n-PDI could be incorporated into PSS and PDAC polyelectrolyte
films by the standard dipping LBL technique.
Unlike many other cases of dye
incorporation into polyelectrolyte films, the PDI moieties were not stripped by
subsequent polyelectrolyte depositions.
Each PDI layer (with variable numbers of
separation layers) is evidently able to block excess deposition of PDI in subsequent
depositions.
The evidence for this is presented in Figure 5.9 and Table 5.2. For
PSS/PDAC “precursor films’ prepared without PDI the amount of PDI uptake increases
with the dipping time to a limiting value that increases with the number of separation
layers (Figure 5.1 and Figure 5.3) (that is, the thicker the film the more PDI can be
adsorbed). However for LBL films in which PDI is absorbed with variable numbers of
separation layers (SL) there is essentially no difference in the PDI uptake (see Table5.2)
suggesting that new PDI cannot intrude on the volume of the polyelectrolyte previously
saturated by PDI. Increasing the number of SL increases the fluorescence intensity, as
seen by the smaller quenching factor in Table 5.2. This effect is also illustrated in
Figure 5.9, which shows the fluorescence intensity increasing steadily with the number of
separation layers but the amount of PDI increases linearly with the number of PDI
depositions.
We wished to study interlayer energy transfer by depositing a good energy
acceptor dye (brilliant green, BG) on the outer layer of the LBL film assembly, but we
found that BG (a dye of moderate size and a net charge of +1) is able to penetrate well
past the outer film surface.
Furthermore we found a very large difference in the
146
effectiveness of BG in quenching the fluorescence of p-PDI and n-PDI, despite the very
similar R0 values (4.7 and 5.2 nm for p-PDI/BG and n-PDI/BG respectively). The
cationic BG is expected to reside preferentially in regions of high PSS density and since
PSS is the layer that surrounds (and interpenetrates) the p-PDI, we propose that the BG
and p-PDI are brought into juxtaposition, leading to very efficient FRET. The n-PDI
layer will be surrounded on both sides by PDAC, which is expected to exclude BG with
the result that approximately 100x more BG is required to produce approximately the
same quenching effect as for p-PDI.
These ideas are illustrated in Figure 5.14.
Therefore the plot of the efficiency of energy transfer vs. the number of layers (Figure
5.13) does not allow us to differentiate between the penetration of BG into the film and
inter-layer energy transfer.
Water-soluble perylene diimide dyes have potential applications as biochemical
dyes16(a),34 and because of their photostability and good fluorescence quantum yield can
be used in single molecule experiments.35 As we have shown here they can be readily
incorporated into polyelectrolyte films by the LBL technique, but we assume that
improvement of their fluorescence quantum yield in such arrays would require even more
steric stabilization and, most likely, a larger number of charged groups per molecular
species.
5.5 References
1. (a) Cormier, R. A., Gregg, B. A. Chem. Mater. 1998, 10, 1309. (b) Würthner, F.,
Thalacker, C., Sautter, A. Adv. Mater. 1999, 11, 754. (c) Antunes, P. A.,
Constantino, C. J. L., Aroca, R. F., Duff, J. Langmuir 2001, 17, 2985. (d) Klebe,
G., Graser, F., Hädicke, E., Berndt, J. Acta Crystallogr. 1989, B45, 69.
147
2. (a) Decher, G., Hong, J. Makromol. Chem. Symp. 1991, 46, 321. (b) Decher, G.
Science 1997, 277, 1232.
3. (a) Wang, Y., Tang, Z. Y., Correa-Duarte, M. A., Liz-Marza’n, L. M., Kotov, N.
A. J. Am. Chem. Soc. 2003, 125, 2830. (b) Wu, A., Yoo, J.K., Rubner, M.F. J.
Am. Chem. Soc. 1999, 121, 4883. (c) Hiller, L., Mendelsohn, J. D., Rubner, M. F.
Nat. Mater. 2002, 1, 59. (d) Tang, Z., Wang, Y., Kotov, N. A. Langmuir 2002, 18,
7035. (e) Park, J., Hammond, P.T. Adv. Mater. 2004, 6, 520. (f) Caruso, F.,
Caruso, R. A., Möhwald, H. Chem. Mater. 2001, 13, 1076. (g) Harris, J. L., Stair,
J.L., Bruening, M.L. Chem. Mater. 2000, 12, 1941.
4. (a) Tedeschi, C., Caruso, F., Möhwald, H., Kirstein, S. J. Am. Chem. Soc. 2001,
122, 5841. (b) Caruso, F., Lichtenfeld, H., Donath, E., Möhwald, H.
Macromolecules 1999, 32, 2317.
5. (a) Birks, J.B., “Photophysics of Aromatic Molecules”, Wiley-Interscience 1971.
(b) Lakowicz, J.R. “principles of Fluorescence Spectroscopy”, 2nd edn., Kluwer
Academic/Plenum Publishers, New York, 1999.
6. (a) Würthner, F., Thalacker, C., Diele, S., Tschierske, C. Chem.-Eur. J. 2001, 7,
2245. (b) Syamakumari, A., Schening, A., Meijer, E.W. Chem.-Eur. J. 2002, 8,
3353.
7. Park. J., Hammond, P.T. Adv. Mater.2005, 16, 520.
8. Ford, W. E. J. Photochem. 1987, 37, 189.
9. Dai, Z., Dähne, L., Donath, E., Möhwald, H. J. Phys. Chem. B. 2002, 106, 11501.
10. Kellog, G. J., Mayes, A. M., Stockton,
Langmuir 1996, 12, 5190.
148
W.B., Ferreira, M., Rubner, M. F.
0.04
Abs. at 580nm
a
0.03
n=1
n=3
0.02
n=2
0.01
0
0
2
4
6
8
10
Dipping Time(Min)
12
14
16
Fluo. Inten.(a.u.)
b
n=1
0
2
4
6
n=2
8
10
12
Dipping Time(Min)
n=3
14
16
Figure 5.1 (a) Absorbance at 580nm and (b) fluorescence intensity of n-PDI as a
function of dipping time for “precursor” LBL films of PEI/(PSS/PDAC)n (n=1, 2, and 3).
149
Absorbance
0.04
a
0.03
0.02
0.01
0
340
400
460
520
580
640
Wavelength(nm)
700
760
Fluo. Inten.
b
560
600
640
680
Wavelength(nm)
720
760
800
Figure 5.2 (a) and (b) Absorption and fluorescence emission spectra of n-PDI dipping
into the preassembled PEI/(PSS/PDAC)6 as a function of immersing time.
150
Abs at 587nm
0.016
a
0.012
0.008
0.004
n=0
n=1
n=2
0
0
1
2
3
4
5 6 7 8 9
Dipping Time(min)
10 11 12 13 14 15
Fluo. Inten.
b
n=0
0
1
2
3
4
5
n=1
n=2
6 7 8 9 10 11 12 13 14 15
Dipping Time(Min)
Figure 5.3 (a) Absorbance at 587nm and (b) fluorescence intensity of p-PDI as a function
of dipping time for “precursor” LBL films of PEI/(PSS/PDAC)n-1/PSS (n=1, 2, and 30).
151
a
Absorbance
0.06
n-PDI
0.06
0.04
0.02
0.04
0
0
0.02
1
2
3
4
5
0
340
440
540
640
740
Wavelength(nm)
0.06
940
0.06
b
Absorbance
840
p-PDI
0.04
0.04
0.02
0
0
0.02
1
2
3
4
5
0
340
440
540
640
740
Wavelength(nm)
840
940
Figure 5.4 (a) and (b) Absorption spectra of n-PDI and p-PDI layer-by-layer assembly
with PSS and PDAC solutions made in 0.5 M NaCl and 3 SLs, respectively. Black
spectra are for the PDI layer as the outside layers and red when the following double
layer of either PSS/PDAC or PDAC/PSS is added with insets showing the absorbance as
a function of the number PDI layers.
152
Fluo. Inten.
4.E+05
a
3.E+05
n-PDI
2.E+05
0
1
2
3
4
5
1.E+05
0.E+00
560
600
640
680
720
Wavelength(nm)
760
800
1.E+06
b
Fluo. Inten.
8.E+05
p-PDI
6.E+05
4.E+05
0
1
2
3
4
5
2.E+05
0.E+00
560
600
640
680
720
760
800
Wavelength(nm)
Figure 5.5 (a) and (b) Fluorescence emission spectra of n-PDI and p-PDI with PSS and
PDAC (0.5M NaCl) LBL films. Black spectra indicate the emission spectra when PDI
layer was formed as the outside layers and red when the following double layer of either
PSS/PDAC or PDAC/PSS formed with insets showing the fluorescence intensity as a
function of the number PDI layers.
153
Fluo. Inten.
a
p-PDI
PSS
PDAC
PSS
PDAC
PSS
Fluo. Inten.
b
n-PDI
PDAC
PSS
PDAC
PSS
PDAC
Figure 5.6 (a) Fluorescence intensity as a function of sequential deposition of
polyelectrolyte layers following the third p-PDI deposition (b) Fluorescence intensity as a
function of sequential deposition of polyelectrolyte layers following the third n-PDI
deposition. All the depositions were performed with 5SLs and 0.5M NaCl in the
polyelectrolyte solutions.
154
0.02
Absorbance
0.02
a
0.016
n-PDI
0.012
0.01
0
0.008
0
1
2
3
4
5
820 880
940
0.004
0
340 400
460 520
580 640
700 760
Wavelength(nm)
0.02
0.02
0.015
Absorbance
0.015
b
0.01
0.005
p-PDI
0.01
0
0
1
2
3
4
5
0.005
0
340
400
460
520 580 640 700
Wavelength(nm)
760
820
880
940
Figure 5.7 (a) and (b) Absorption spectra of n-PDI and p-PDI with PSS and PDAC
(3SLs) in the LBL films (with no NaCl in the polyelectrolyte solution). Black spectra
indicate the absorption or emission spectra when PDI layer was formed as the outside
layer and red when the following double layer of either PSS/PDAC or PDAC/PSS was
deposited. The insets show the absorbance at 580nm as a function of the number PDI
layers before and after deposition of double layer onto the PDI layer.
155
Fluo. Inten
a
n-PDI
0
560
600
640
680
1
2
720
760
Wavelength(nm)
3
800
4
840
5
880
Fluo. Inten
b
p-PDI
0
560
600
640
1
680
720
760
Wavelength(nm)
2
3
800
4
840
5
880
Figure 5.8 (a) and (b) fluorescence spectra for n-PDI and p-PDI. Black spectra indicate
the absorption or emission spectra when PDI layer was formed as the outside layer and
red when the following double layer of either PSS/PDAC or PDAC/PSS was deposited.
The insets show the or fluorescence intensity as a function of the number PDI layers.
156
Abs. at 577nm
1
0.06
0.8
y = 0.0107x
0.04
0.02
3
0.6
0.4
7
5
9
1
0.2
Norm. Fluo. Inten.
1.2
0.08
0
0
0
1
2
3
4
5
Number of n-PDI Layers
6
7
Figure 5.9 Absorption vs. the number of n-PDI layers deposited (●) and the normalized
fluorescence intensity (■) for each layer. The number of separation layers are indicated
between each n-PDI deposition (SL=1, 3, 5, 7, 9).
157
a
b
Figure 5.10 TM-AFM angle view images of (a) before p-PDI deposition and (b) after pPDI depositions.
158
Blue: PSS
Black: PDAC
Red: PDI
Intralayer-FRET Interlayer-FRET Intralayer-FRET
Figure 5.11 Schematic representation of p-PDI in LBL films with 3SLs along with
showing the intralayer and interlayer FRET process.
159
0.8
ET Eff.
0.6
0.4
a
n-PDI
0.2
0
0
0.001 0.002 0.003 0.004 0.005 0.006 0.007
a
Spin-Coating Concentration of BG(M)
1.2
ET Eff.
1
0.8
0.6
b
p-PDI
0.4
0.2
0
0
0.0001 0.0002 0.0003 0.0004 0.0005 0.0006
Spin Coating Concentration of [BG]/M
Figure 5.12 (a) and (b) Energy transfer efficiency as a function of spin-casting BG
concentration for 4 n-PDI and p-PDI LBL films with 3SLs.
160
1
ET Eff.
0.8
a
n-PDI
0.6
0.4
0.2
0
2
3
4
5
6 7
8 9 10 11 12 13
Number of n-PDI layers
ET Eff.
0.8
p-PDI
b
0.6
0.4
0.2
0
2
3
4
5 6 7 8 9 10 11 12 13
Number of p-PDI Layers
Figure 5.13 (a) Energy transfer efficiency as a function of number of n-PDI layers at BG
concentration of 1.3x10-3M. (b) Energy transfer efficiency as a function of number of pPDI layers at BG concentration of 8.3x10-6M.
161
(p-PDI/3SL)n/p-PDI/PSS/PDAC/PSS/BG
Red: p-PDI or n-PDI
Blue: PSS Layer
Black: PDAC Layer
Green: BG Layer
(n-PDI/3SL)n/n-PDI/PDAC/BG
Figure 5.14 Schematic representation of BG distribution in p-PDI and n-PDI LBL films
with 3SLs.
162
Table 5.1 FRET parameters for D-A pairs of p-PDI, n-PDI and BG
D-A
p-PDI/p-PDI
n-PDI/n-PDI
p-PDI/BG
n-PDI/BG
J
7.99x1014
8.37x1014
4.03x1015
5.22x1015
R0(nm)
3.58
4.50
4.68
5.18
163
C0(mM)
9.79
4.93
4.32
3.25
Table 5.2 Characteristics of UV-Vis and fluorescence spectra of p-PDI and n-PDI LBL films with 1, 3, and 5 SLs with and
without using NaCl.
p-PDI/NaCla
n-PDI/NaCla
p-PDI
n-PDI
SLs
Abs.b
(x1000)
Peakc
1
3
5
1
3
5
1
3
5
1
3
5
9.6
10.9
9.8
2.6
2.8
2.8
11.2
11.5
11.6
3.4
3.1
3.0
587
592
588
588
588
588
575
576
577
575
575
575
Extractiond
N
N
N
Y
Y
Y
N
N
N
Y
Y
Y
Fluo. Peakc
621
620
621
619
620
620
612
614
613
613
615
614
Quen. Factor.e
35
25
9
15
19
8
43
33
22
17
14
12
a. 0.5 M NaCl was used for the polyelectrolyte solutions.
b. Slopes of absorbance around 580 nm from either p-PDI or n-PDI as outside layer versus the number of PDI layers;
c. S0ÆS1 UV-Vis peak position or S1ÆS0 fluorescence emission peak position for either p-PDI or n-PDI as the outmost layers;
d. N means no extraction and Y means having extraction;
e. Quenching factors obtained by comparing the fluorescence intensity of the p-PDI and n-PDI aqueous solutions in thin cells with p- or nPDI LBL films.
164
Chapter 6 Some Preliminary Results and Suggestions for Future work
Two projects were initiated as part of this Ph. D. research which suggests some
new and interesting directions for future work with PDI materials. These are briefly
described in the following sections.
6.1 Synthesis of n-type and p-type PDI materials
6.1.1 Introduction
The drive towards inexpensive electronic technologies involving organic thin-film
field-effect transistors (OTFTs) has resulted in part from increasing demand for pervasive
computing devices that will require light weight construction and inexpensive
components for memory, logic, and displays. Moreover, OTFTs are well known to have
many advantages such as large-area coverage, structural flexibility, shock resistance, and
low-temperature processing, in comparison with the conventional silicon based TFTs.1
Historically, OTFTs based on p-type materials such as pentacene, poly(3-hexylthiophene)
or oligothiophenes in which holes are the majority carriers, have received most of the
attention. The highest hole mobility for pentacene TFTs was 1.5 cm2V-1s-1.2 Recently, the
hole mobility had been reported to reach 8-20cm2V-1s-1 in rubrene single crystal TFTs.3
High performance TFTs based on n-type materials in which electrons are the majority
carriers, are also desired since they will enable the fabrication of complimentary circuits.
However, the performance of the best n-type materials is not quite as good as the best ptype materials, although there has been increased attention to n-type materials such as
phthalocyanine and fluorinated polythiophene with the corresponding improvements in
the field effect mobility, the on-to-off current ratio and the threshold voltage.4 Perylene
diimides are good candidates for n-type conductors because (1) they have relatively large
165
electron affinities, (2) they typically assemble in π-stacks in the solid state which
enhances the intermolecular π-orbital overlap and facilitates charge transport, and (3) the
nature of π-orbital interactions can be tuned with high precision by changing the
substituents on the imide N atoms, offering the opportunity to optimize transport
properties by systematic structure variations, which in turn influences the crystal packing.
Until now, the highest mobility of PDIs can reach to 2 cm2V-1s-1 with on-to-off current
ratios of 107.5
Although both p-type and n-type materials with very good properties for
the fabrication of TFTs have been independently discovered, there has not be any report
on materials which can have the n-type and p-type properties in one molecule. In this
research we are trying to combine the perylene diimides and fused benzene ring
molecules such as naphthalene and phenanthrene in order to obtain materials which can
show both the n-type and p-type field effect. In this work we introduce the naphthalene or
phenanthrene groups in the bay area and not on the imide head-tail positions. The double
substitutions in the bay area of PDIs are expected to impose less twisting of the perylene
core compared with the tetra substituents in the bay region.
6.1.2 Synthesis
The synthetic routes for the PDI compounds are shown in Figure 6.1, which
involves three steps.
6.1.2.1
Bromination of Perylene Dianhydride(PDA)
The bromination of PDA was repeated according to the patent procedure as
shown in the first step in Figure 6.1.6 A mixture of 31.3 (80.0mmol) of PDA and 480g of
100% sulfuric acid was stirred for 12hrs at room temperature, and subsequently I2 (0.77g,
3.0mmol) was added. The reaction mixture was heated to 85ºC, and 29g (177mmol) of
166
bromine was added dropwise over a time period of 8hrs. After bromine addition, the
reaction mixture was heated for an additional 10 hrs at 85ºC and cooled to room
temperature. The excess bromine was removed by a gentle stream of N2 gas and 65 mL
of water was added carefully. The resulting precipitate was separated by filtration
through a G4 funnel, washed with 300g of 86% sulfuric acid and a large amount of water,
and dried in a vacuum to give 37.6g (87%) of red powder. The crude product (written as
PDA-Br2) could not be purified since it is insoluble in organic solvents. PDA-Br2 was
used for the subsequent imidization with cyclohexylamine.
6.1.2.2
Imidization of crude Bromination Product7
A suspension of PDA-Br2 (0.95g, 1.72mmol) obtained in the above reaction,
cyclohexylamine (0.502g, 5.07mmol), and acetic acid (0.50g, 8.33mmol) in 30mL of Nmethyl-2-pyrrolidinone was stirred at 85ºC under Argon gas protection for 6 hrs. After
mixture was cooled to room temperature, the solution was poured into 300mL of 1M HCl
for precipitation overnight. The precipitate was separated by filtration, washed with
100mL of MeOH, and dried in a vacuum. The crude product was purified by silica gel
column chromotography with CH2Cl2 as eluent. This product is denoted as PDI-Br2. This
product was characterized by both TLC and Mass Spectroscopy for purification.
6.1.2.2
Nucleophilic Replacement of Bromine atoms8
PDI-Br2 (0.90g, 1.3mmol), 1-Naphthanol (1.5g, 10.4mmol) and K2CO3 (0.3g,
2.2mmol) were heated in 50mL of NMP at 85ºC overnight under argon gas protection.
After cooling to room temperature, the reaction mixture was poured into 150mL of 1M
HCl for precipitation. The crude product was separated by vacuum filtration. Pure
product of PDI-NP1 was obtained by running silicon gel column chromatography with
167
CH2Cl2/MeOH (9:1) as the eluent. Its mass spectroscopy and H-NMR were shown in
Figure 6.2a and b respectively. The perylene core chemical shift resides above 8ppm and
naphthalene around 6 and 8ppm. They are both multipeaks. CDCl3 was used for H-NMR
measurements.
Following the same protocol as above, the pure product of PDI-NP2 and PDI-PN9
were synthesized. Similar Mass Spectroscopy and H-NMR with PDI-NP1 were obtained
for PDI-NP2, as shown in Figure 6.3. The mass spectroscopy for PDI-PN9 was shown in
Figure 6.4 with a very strong molecular ion peak.
The chemical structures of these 3 PDIs are shown in Figure 6.5.
6.1.3 Photophysics of the synthesized PDIs
The normalized absorption and emission spectra of PDI-NP1, PDI-NP2 and PDIPN9 in DCE are shown in Figure 6.6. The absorption spectra for PDI-NP1 and PDI-NP2
are very similar with three absorption peaks that are typical for a PDI core except there is
a 4nm red shift for PDI-NP1. This red shift is attributed to the higher steric hindrance
originating from the 1-naphthanol substituent. Comparing to PDI-NP1 and PDI-NP2,
there is stronger S0,0-S1,1 absorption peak observed for PDI-PN9, which might be
attributed to dimer formation for PDI-PN9 in DCE due to the larger π-π interactions from
phenanthrene in the bay region.9 Emission spectra at Ex=505nm are shown in Figure
6.6b. A single emission peak at 567nm was observed for PDI-NP1 and PDI-NP2, which
is typical for bay area substituted PDIs. The same position peak for PDI-NP1 and PDINP2 indicates no significant coupling and interaction between naphthalene and PDI core
is present. PDI-NP2 has a higher fluorescence quantum yield of 0.65 and PDI-NP1 has a
quantum yield about 0.11. The lower quantum yield of PDI-NP1 is due to the larger steric
168
hindrance from 1-naphthanol substituent leading to more twisting of the PDI core. The
Stokes shifts are 25 and 29nm for PDI-NP1 and PDI-NP2. Two emission peaks are
located a 573nm and 620nm observed for PDI-PN9, leading to a Stokes shift of 32nm.
The quantum yield is about 0.01 for PDI-PN9 presumably due to aggregation or the
strong coupling between phenanthrene and PDI core.
Emission spectra with the excitation wavelength chosen for the bay region
substituents are shown in Figure 6.6c. Typical naphthalene emission spectra were
observed for PDI-NP1 and PDI-NP2. The excitation for the phenanthrene substituent also
resulted in the emission from phenanthrene alone. All these results indicate that no
coupling exists between perylene core and bay substituent. In other words, the bay
substituent chromophore and perylene core behave independently. Further investigation
of the electrochemical properties of these compounds will provide us more information
about this point.
6.1.4. Future work
The decomposition temperature for PDI-NP1 and PDI-NP2 is around 410ºC. DSC
measurements have shown a Tg of 277 ºC of PDI-NP1. The Tg for PDI-NP2 is not very
obvious. This indicated the good thermostability of synthesized PDIs. Future research of
using these synthesized for the TFT study will be conducted to characterize the electron
and hole carrier mobility.
6.2 Self-assembly of n-type semiconductor PDIs
6.2.1 Introduction
One-dimensional nanostructures (such as nanowires) of semiconductor materials
have attracted increased interest in recent years due to their promising applications in
169
optical and electronic nanodevices.10 However, most nanowires reported to date are based
on the inorganic materials. A few one-dimensional nanostructures (e.g. nanofibers) have
been fabricated with conducting polymers such as polyaniline and polyacetylene, but the
crystalline structure of polymer nanofibers is often difficult to control due to the
complicated intermolecular interaction.11 Recently, some works have proven that selfassembly of through strong π-π stacking is an effective approach to one-dimension
nanostructures for aromatic molecules, particularly the larger macrocyclic aromatic
molecules such as hexabenzocoronene.12 Very recently, Zang Ling et al. used “phase
transfer” self-assembly between good and poor solvents and a surface annealing based
self-assembly method for making organized nanobelts of PDIs by controlling the π-π
stacking and the hydrophobic interactions between the side chains linked at the imide
positions.13-15 This implies a potential method to enhance the charge carrier mobility,
which is believed be favored along the π-π stacking direction. Utilizing the solution based
“phase transfer” self-assembly method, nanobelt or even nanotubes can be obtained with
our synthesized PDIs.
6.2.2 Self-assembly of PDIs
6.2.2.1 Chemical structures of PDIs
The synthesis of PDIs was described in Chapter 2. These PDIs have branched
chains originating from a tertiary nitrogen (their Chemical Structures are also shown in
Figure 6.5). These three PDIs studied here are denoted as PDI-DMPA, PDI-DEPA and
PDI-DMEA. All three have good solubility in chloroform and very low solubility in
hexane.
6.2.2.2 Solution based Self-assemblying
170
The solution based self-assembly of PDIs was performed using a so-called “phase
transfer” method.13 Briefly, the nanostructure growth of the PDIs assembly was acheived
through slow crystallization at the interface between a “good” and a “poor” solvent,
where the slow “phase transfer” between the two solvents decreases the solubility at the
interface. The poor solvent (e.g., hexane) is normally quite different (e.g., in term of
polarity and density) from the good solvent (e.g. chloroform), thus providing the
possibility to keep the two solvents in separate phases for an extended period (longer than
one day). Typically, a larger amount (15:1, v/v) of hexane was carefully transferred atop
a concentrated chloroform solution of PDI solutions (~0.1mM) in a test tube. Within 20
minutes, red crystals formed at the interface, followed by very slow diffusion into the
upper phase of hexane. The nanobelts can be transferred into the hexane for dilution by
pipeting. Then the suspension was drop-cast onto the clean silicon wafer. The images
were then measured by SEM.
Scanning Electron Microscopy (SEM) images were taken on a LEO 1530
scanning electron microscope, which has a GEMINI field emission column with a
thermal field emitter. The voltage was set at 10 kV. The samples were prepared by
casting a hexane solution of PDI nanoassemblies onto a clean silicon wafer, followed by
drying in air.
Typical images of SEMs for PDI-DEPA are shown in Figure 6.7. Nanobelts were
formed for PDI-DEPA with >10µm in length and 400nm in width, which is around the
same size as Ling Zang et al. obtained for their PDIs. Large area nanobelt piles were
found over the surfaces for PDI-DMPA shown in Figure 6.8, and they showed quite
171
narrow size (width) distribution. The π-π stacking was found to be along the nanobelt
optical axis by Ling Zang et al. 14
SEM images of PDI-DMEA nanotubes are shown in Figure 6.9. Figure 6.11a
showed a chunk of broken nanotubes. A wall thickness of 500nm was obtained for the
nanotube from the Figure 6.9b and c. Evidently the smaller size of the side chains of PDIDMEA make it possible to form nanotubes. The nanotube structures are very fragile as
shown in Figure 6.9a. As far as we know, this is the first report of nanotube synthesis
from organic semiconductors.
6.2.3 Future of Self-assembly of PDIs
The PDIs we synthesized have a tertiary amine in the side chains, which is
different from the PDI molecules that Ling Zang studied. Therefore, the protonation of
tertiary amine by dicarboxylic acid molecules such as butanedioic acid or hexanedioic
acid will lead to the crosslinking of the PDI side chains in the nanoassemblies. The
structural stability upon diacid crosslinking can be tested by dissolving the
nanoassemblies into chloroform to see if the nanostructure can be retained or not. In the
same way, the crosslinking of diamines by di-bromo agents could also be examined.
We have prepared a fairly large number of compounds of this class, with chemically
distinct branched structures (see Chapter 2), which may modify the kinds of linear
structures that can be formed by the phase-transfer or related methods.
6.3 References
1. (a) Dimitrakopoulos, C. D., Mascaro, D. J. IBM J. Res. Dev. 2001, 45, 1. (b)
Dimitrakopoulos, C. D., Malenfant, P. R. L. Adv. Mater. 2002, 14, 99.
172
2. Lin, Y.-Y., Gundlach, D. J., Nelson, S. F., Jackson, T. N. IEEE Electron Device
Lett. 1997, 18, 606.
3. (a) Sundar, V. C., Zaumseil, J., Podzorov, V., Menard, E., Willett, R. L., Someya,
T., Gershenson, M. E., Rogers, J. A. Science 2004, 303, 1644. (b) Podzorov, V.,
Menard, E., Borissov, A., Kiryukhin, V., Rogers, J. A., Gershenson, M. E. Phys.
Rev. Lett. 2004, 93, 086602.
4. (a) Bao, Z., Lovinger, A. J., Brown, J. J. Am. Chem. Soc. 1998, 120, 207. (b) Bao.
Z. Adv. Mater. 2000, 12, 227. (c) Facchetti, A., Mushrush, M., Katz, H. E.,
Marks, T. J. Adv. Mater. 2003, 15, 33.
5. Chesterfield, R. J., Mckeen, J. C., Newman, C. R., Ewbank, P. C., da Silva Filho,
D. A., Brédas, J-L., Miller, L. L., Mann, K. R., Frisbie, C. D. J. Phys. Chem.B
2004, 108, 19281-19292.
6. Böhm, A., Arms, H., Henning, G., Blaschka, P. (BASF AG) German Pat. DE
19547209A1, 1997. Chem. Abstr. 1997, 127, 96569g.
7. Würthner, F., Stepanenka, V., Chen, Z., Saha-Möller, C. R., Kocher, N., Stalke,
D. J. Org. Chem. 2004, 69, 7933.
8. (a) Qu, J., Kohl, C., Pottek, M., Müllen, K. Angew. Chem., Int. Ed. 2004, 43,
1528. (b) Kohl, C., Weil, T., Qu, J., Müllen, K. Chem.-Eur. J. 2004, 10, 5297.
9. van der Boom, T., Evmeneko, G., Dutta, P., Wasielewski, M. R. Chem. Mater.
2003, 15, 4068.
10. Law, M., Goldberger, J., Yang, P. Annu. Rev. Mater. Res. 2004, 34, 83.
11. (a) Huang, J. X., Virji, S., Weiller, B. H., Kaner, R. B. J. Am. Chem. Soc. 2003,
125, 314. (b) Lee, H. J., et al. J. Am. Chem. Soc. 2004, 126, 16722.
173
12. Kastler, M., Pisula, W., Wasserfaller, D., Pakula, T., Müllen, K. J. Am. Chem.
Soc. 2005, 127, 4286.
13. Balakrishnan, K., Datar, A., Naddo, T., Huang, J., Oikter, R., Yen, M., Zhao, J.,
Zang, L. J. Am. Chem. Soc. 2006, ASAP.
14. Balakrishnan, K., Datar, A., Oikter, R., Chen, H., Zuo, J., Zang, L. J. Am. Chem.
Soc. 2005, 127, 10496.
15. Datar, A., Oikter, R., Zang, L. Chem. Comm. 2006, 1649.
174
Br
O
O
O
O
i: Br2, 100% H2SO4/I2
O
80oC, 12h, 87%
O
O
O
O
O
O
O
Br
ArOH=1-Naphthalenol PDI-NP1
2-Naphthalenol PDI-NP2
9-Phenanthrol PDI-PN9
ii: Cyclohexylamine, NMP, 85oC, overnight
Ar
O
Br
O
O
C6H11 N
N C6H11
O
Ar-OH, K2CO3
O
O
iii:NMP, 85oC,
C6H11 N
N C6H11
O
O
O
O
Br
Ar
Figure 6.1 Synthetic routes to Compounds PDI-NP1, PDI-NP2, and PDI-PN9
175
a
b
Figure 6.2 Mass Spectrum of PDI-NP1 by CI+ (a) and H-NMR in CDCl3 (b).
176
a
b
Figure 6.3 Mass Spectrum of PDI-NP2 by CI- (a) and H-NMR in CDCl3 (b).
177
Figure 6.4 Mass Spectrum of PDI-PN9 by CI+.
178
O
O
O
O
O
O
O
O
O
N
N
N
N
N
N
O
O
O
O
O
O
O
PDI-NP1
O
PDI-NP2
O
H3 C
H3 C
PDI-PN9
O
NH2CH2CH2C N
CH3
N CH2CH2CH2N
O
CH3
O
PDI-MAPA
O
H3CH2C
H3CH2C
O
NH2CH2CH2C N
N CH2CH2CH2N
O
CH2CH3
CH2CH3
O
PDI-DEPA
O
H3C
H3 C
O
NH2CH2C N
N CH2CH2N
O
O
PDI-DMEA
Figure 6.5 Chemical Structures of Synthesized PDIs
179
O
CH3
CH3
a
Absorbance
1.2
0.8
0.4
0
220
340
400
460 520 580
Wavelength(nm)
640
700
b
Norm. Fluo. Inten.
0.016
0.8
0.012
0.6
0.4
0.008
0.2
0.004
0
0
520
570
1
Norm. Fluo. Inten.
760
0.02
1
Norm. Fluo. Inten.
280
620
670
720
Wavelength(nm)
770
c
0.8
0.6
0.4
0.2
0
270
300
330
360
390
420
Wavelength(nm)
450
480
Figure 6.6 Normalized absorption (a) and emission spectra (b and c) of PDI-NP1 (Red),
PDI-NP-2(pink) and PDI-PN9 (Blue) in DCE. Emission spectra in Figure 6.7b were
taken at Ex=505nm. Emission spectra in Figure 6.7c were taken at Ex=260nm for PDINP1 and PDI-NP2 and Ex=290nm.
180
Figure 6.7 SEM images of PDI-DEAPA.
181
Figure 6.8 Large area SEM images of PDI-DMPA
182
Figure 3.9 SEMs images of PDI-DMEA.
183
Bibliography
1. Hunter, C. A., Sanderes, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525.
2. Cao, T., Wei, L., Cao, W. Langmuir 2002, 18, 750.
3. Wei, X-L., Wang, Y. Z., Long, S. M., Bobeczko, C., Epstein J. Am. Chem. Soc.
1996, 118, 2545.
4. Tomiki, I. J. Mater.Chem. 2003, 13, 2037.
5. Liu, S., Sui, G., Cormier, R. A., Leblane, R. M., Gregga, B. A. J. Phys. Chem. B.
2002, 106, 1307
6. Horowitz, G. Adv. Mater. 1998, 10, 365.
7. Nueteboom, E. E., Meskers, S. C. J., Meijer, E. W., Janssen, R. A. Macromol.
Chem. Phys. 2004, 205, 217.
8. Liu, Y., Wang, N., Li, Y., Liu, H., Li, Y., Xiao, J., Xu, He., Xu, X., Huang, C.,
Cui, S., Zhu, D. Macromolecules 2005, 38, 4880.
9. Köhler, K., Shchukin, D. G., Sukhorukov., G. B., Möhwald, H. Macromolecules
2004, 37, 9546.
10. Gao, C., Wang, B., Feng, J., Shen, J. Macrmolecules 2004, 37, 8836.
11. Zhao, L., Cebeci, F. C., Cohen, R., Rubner, M. Nano. Lett. 2004, 7, 1349.
12. Liu, N., Dunphy, D., Atanassov, P., Bunge, S. D., Chen, Z., Lopez, G. P., Boyle,
T. J., Brinker, C. J. Nano. Lett. 2004, 7, 551.
13. Dante, S., Advincula, R., Frank, C. W., Stroeve, P. Langmuir 1999, 15, 193.
14. Cho, J., Quinn, J. F., Caruso, F. J. Am. Chem. Soc. 2004, 126, 2270.
184
15. Chiu, J. J., Kim, B. J., Kramer, E. J., Pine, D. J. J. Am. Chem. Soc. 2005, 127,
5036.
16. Lee, M., Cho, B. K., Zin, W. C. Chem. Rev. 2001, 101, 3869.
17. Klok, H-A., Lecommandous, S. Adv. Mater. 2001, 13, 1217.
18. Lazzari, M., Lopez-Quintela, M. Adv. Mater. 2003, 15, 1583.
19. Ikkala, O., Brinke, G. T. Science 2002, 295, 29, 2407.
20. Park, C., Yoon, J., Thomas, E. L. Polymer 2003, 44, 6276.
21. Beijonne, D., Pourtois, G., Hennebicq, E., Herz, L. M., Friend, R. H., Scholes, G.
D., Setayesh, S., Müllen, K., Brédas, J. L. PNAS 2002, 99, 10982.
185
Vita
Tingji Tang was born to Jiaen Tang and Chengfeng Lu on May 11, 1977 in
Nanluo village, Penglai City, Shandong Province, P.R. China. After completing his work
at 2nd place in Xujiaji Middle School, he went to Penglai No. 1 High School in 1992. Due
to his strong performance in the college entrance examination in 1995, he entered
Qingdao Institute of Chemical technology (now as Qingdao University of Science and
Tehnology) and earned his Bachelor of Engineering degree in Rubber Science and
Engineering in 1999. Then he continued to pursue his Master Degree of Science in
Polymer Chemistry and Physics from Peking University in 2002. In fall, 2002, he came
to the United States and entered the Graduate School of the University of Texas at
Austin. He was appointed as Graduate Teaching and Research Assistant in the course of
his graduate studies. He married Yanbo Wang In 2002 and had a son, Bryan Benshuo
Tang, in 2005.
Permanent Addess: Nanoluo Village, Beigou Town, Penglai City, Shandong Province, P.
R. China, 265602.
This dissertation was typed by the author.
186